Editing domains of Trypanosoma brucei mitochondrial RNAs identified by secondary structure

High-throughput screening of compounds targeting RNA editing in Trypanosoma brucei: Novel molecular scaffolds with broad trypanocidal effects

Mitochondrial uridine insertion/deletion RNA editing, catalyzed by a multiprotein complex (editosome), is essential for gene expression in trypanosomes and Leishmania parasites. As this process is absent in the human host, a drug targeting this mechanism promises high selectivity and reduced toxicity. Here, we successfully miniaturized our FRET-based full-round RNA editing assay, which replicates the complete RNA editing process, adapting it into a 1536-well format. Leveraging this assay, we screened over 100,000 compounds against purified editosomes derived from Trypanosoma brucei, identifying seven confirmed primary hits. We sourced and evaluated various analogs to enhance the inhibitory and parasiticidal effects of these primary hits. In combination with secondary assays, our compounds marked inhibition of essential catalytic activities, including the RNA editing ligase and interactions of editosome proteins. Although the primary hits did not exhibit any growth inhibitory effect on parasites, we describe eight analog compounds capable of effectively killing T. brucei and/or Leishmania donovani parasites within a low micromolar concentration. Whether parasite killing is - at least in part - due to inhibition of RNA editing in vivo remains to be assessed. Our findings introduce novel molecular scaffolds with the potential for broad antitrypanosomal effects.

Hammerhead ribozyme-based U-insertion and deletion RNA editing assays for multiplexing in HTS applications

Untranslatable mitochondrial transcripts in kinetoplastids are decrypted post-transcriptionally through an RNA editing process that entails uridine insertion/deletion. This unique stepwise process is mediated by the editosome, a multiprotein complex that is a validated drug target of considerable interest in addressing the unmet medical needs for kinetoplastid diseases. With that objective, several in vitro RNA editing assays have been developed, albeit with limited success in discovering potent inhibitors. This manuscript describes the development of three hammerhead ribozyme (HHR) FRET reporter-based RNA editing assays for precleaved deletion, insertion, and ligation assays that bypass the rate-limiting endonucleolytic cleavage step, providing information on U-deletion, U-insertion, and ligation activities. These assays exhibit higher editing efficiencies in shorter incubation times while requiring significantly less purified editosome and 10,000-fold less ATP than the previously published full round of in vitro RNA editing assay. Moreover, modifications in the reporter ribozyme sequence enable the feasibility of multiplexing a ribozyme-based insertion/deletion editing (RIDE) assay that simultaneously surveils U-insertion and deletion editing suitable for HTS. These assays can be used to find novel chemical compounds with chemotherapeutic applications or as probes for studying the editosome machinery.

Editosome accessory factors KREPB9 and KREPB10 in Trypanosoma brucei

Multiprotein complexes, called editosomes, catalyze the uridine insertion and deletion RNA editing that forms translatable mitochondrial mRNAs in kinetoplastid parasites. We have identified here two new U1-like zinc finger proteins that associate with editosomes and have shown that they are related to KREPB6, KREPB7, and KREPB8, and thus we have named them Kinetoplastid RNA Editing Proteins, KREPB9 and KREPB10. They are conserved and syntenic in trypanosomatids although KREPB10 is absent in Trypanosoma vivax and both are absent in Leishmania. Tandem affinity purification (TAP)-tagged KREPB9 and KREPB10 incorporate into ~20S editosomes and/or subcomplexes thereof and preferentially associate with deletion subcomplexes, as do KREPB6, KREPB7, and KREPB8. KREPB10 also associates with editosomes that are isolated via a chimeric endonuclease, KREN1 in KREPB8 RNA interference (RNAi) cells, or MEAT1. The purified complexes have precleaved editing activities and endonuclease cleavage activity that appears to leave a 5' OH on the 3' product. RNAi knockdowns did not affect growth but resulted in relative reductions of both edited and unedited mitochondrial mRNAs. The similarity of KREPB9 and KREPB10 to KREPB6, KREPB7, and KREPB8 suggests they may be accessory factors that affect editing endonuclease activity and as a consequence may affect mitochondrial mRNA stability. KREPB9 and KREPB10, along with KREPB6, KREPB7, and KREPB8, may enable the endonucleases to discriminate among and accurately cleave hundreds of different editing sites and may be involved in the control of differential editing during the life cycle of T. brucei.

Preferential interaction of a 25kDa protein with an A6 pre-mRNA substrate for RNA editing in Trypanosoma brucei

Mitochondrial gene expression in kinetoplastids is controlled after transcription, potentially at the levels of RNA maturation, stability and translation. Among these processes, RNA editing by U-insertion/deletion catalysed by multi-subunit editing complexes is best characterised at the molecular level. Nevertheless, mitochondrial RNA metabolism overall remains poorly understood, including the potential regulatory factors that may interact with the relevant catalytic molecular machines and/or RNA substrates. Here we report on a approximately 25kDa polypeptide in mitochondrial extracts that exhibits a preferential "zero-distance" photo-crosslinking interaction with an A6 pre-mRNA model substrate for RNA editing containing a single [(32)P] at the first editing site. The approximately 25kDa polypeptide purified away from editosomes upon ion-exchange chromatography and glycerol gradient sedimentation. Competition assays with homologous and heterologous transcripts suggest that the preferential recognition of the A6 substrate is based on relatively low-specificity RNA-protein contacts. Our mapping and substrate truncation analyses suggest that the crosslinking activity primarily targeted a predicted stem-loop region containing the first editing sites. Consistent with the notion that pre-mRNA folding may be required, pre-annealing with guide RNA abolished crosslinking. Interestingly, this preferential protein interaction with the A6 substrate seemed to require adenosine 5'-triphosphate but not hydrolysis. As in other biological systems, fine regulation in vivo may be brought about by transient networks of relatively low-specificity interactions in which multiple auxiliary factors bind to mRNAs and/or editing complexes in unique higher-order assemblies.

TbMP81 is required for RNA editing in Trypanosoma brucei

Most mitochondrial mRNAs are edited in Trypano soma brucei by a series of steps that are catalyzed by a multienzyme complex that is in its initial stages of characterization. RNA interference (RNAi)-mediated repression of the expression of TbMP81, a zinc finger protein component of the complex, inhibited growth of bloodstream and insect forms, and blocked in vivo RNA editing. This repression preferentially inhibited insertion editing compared with deletion editing in vitro. It resulted in reduced specific endoribonucleolytic cleavage and a greater reduction of U addition and associated RNA ligation activities than U removal and associated RNA ligation activities. The repressed cells retained 20S editing complexes with several demonstrable proteins and adenylatable TbMP52 RNA ligase, but adenlyatable TbMP48 was not detected. Elimination of TbMP48 by RNAi repression did not inhibit cell growth or in vivo editing in either bloodstream or procyclic forms. These results indicate that TbMP81 is required for RNA editing and suggest that the editing complex is functionally partitioned.

Interactions of mRNAs and gRNAs involved in trypanosome mitochondrial RNA editing: structure probing of an mRNA bound to its cognate gRNA

Posttranscriptional editing of trypanosome mitochondrial messenger RNA is directed by small guide RNAs (gRNAs). Using crosslinking techniques, we have previously shown that the gRNA base pairs to the mRNA via a 5' anchor, whereas its 3' U-tail interacts with upstream purine-rich mRNA sequences. The incorporation of crosslinking data into RNA folding programs produced similar structure predictions for all gRNA/mRNA pairs examined. This suggests that gRNA/mRNA pairs can form common secondary structure motifs that may be important for recognition by the editing complex. In this study, the structure of CYb mRNA crosslinked to gCYb-558 was examined using solution-probing techniques. The mRNA/gRNA crosslinked molecules are efficient substrates for gRNA-directed cleavage. In addition, when the cleavage assay is performed in the presence or absence of additional UTP, the activities of both the U-specific exonuclease and terminal uridylyl transferase (tutase) can be detected. These results indicate that a partial editing complex can assemble and function on these substrates suggesting that the crosslink captured the molecules in a biologically relevant interaction. The structure probing data directly show that the U-tail protects several mRNA bases predicted to be involved in the U-tail-mRNA duplex. In combination with our previous studies, these new data provide additional support for the predicted secondary structure of interacting gRNA/mRNA pairs.

Uridylate addition and RNA ligation contribute to the specificity of kinetoplastid insertion RNA editing

RNA editing in Trypanosoma brucei inserts and deletes uridylates (U's) in mitochondrial pre-mRNAs under the direction of guide RNAs (gRNAs). We report here the development of a novel in vitro precleaved editing assay and its use to study the gRNA specificity of the U addition and RNA ligation steps in insertion RNA editing. The 5' fragment of substrate RNA accumulated with the number of added U's specified by gRNA, and U addition products with more than the specified number of U's were rare. U addition up to the number specified occurred in the absence of ligation, but accumulation of U addition products was slowed. The 5' fragments with the correct number of added U's were preferentially ligated, apparently by adenylylated RNA ligase since exogenously added ATP was not required and since ligation was eliminated by treatment with pyrophosphate. gRNA-specified U addition was apparent in the absence of ligation when the pre-mRNA immediately upstream of the editing site was single stranded and more so when it was base paired with gRNA. These results suggest that both the U addition and RNA ligation steps contributed to the precision of RNA editing.

Guide RNA requirement for editing-site-specific endonucleolytic cleavage of preedited mRNA by mitochondrial ribonucleoprotein particles in Trypanosoma brucei

RNA editing in trypanosome mitochondria entails the posttranscriptional internal addition and occasional deletion of uridines from precursor mRNAs. Ample evidence exists to show that the information specifying the site and number of uridines added or deleted comes from small, mitochondrially encoded guide RNAs (gRNAs). More recent work indicates that the process involves an enzymatic cascade, initiating with an endonucleolytic cleavage of the pre-mRNA at an editing site. The cleaved editing site can undergo uridine (U) addition to or deletion from the 3' end of the 5' fragment via a mitochondrial terminal uridylyl transferase (TUTase) or terminal uridylyl exonuclease, respectively. Mitochondrial RNA ligase subsequently rejoins the mRNA. Activities to carry out these processes have been found in trypanosome mitochondria, including an editing-site-specific endonuclease activity which cleaves preedited but not edited mRNAs. We have found that this enzymatic activity cosediments with the same 19S ribonucleoprotein particle previously shown to contain TUTase, RNA ligase, and gRNAs and remains stable after salt treatment. Depletion of endogenous cytochrome b gRNAs by the addition of complementary oligonucleotides in vitro completely inhibits editing-site cleavage of synthetic preedited cytochrome b mRNA. The addition of synthetic cognate gRNA for cytochrome b but not unrelated small RNA restores editing-site cleavage. These studies show that in addition to specifying the site and number of uridines added or deleted, gRNAs provide the necessary information for cleavage by the editing-site-specific endonuclease.

Purification of a functional enzymatic editing complex from Trypanosoma brucei mitochondria

Kinetoplastid mitochondrial RNA editing, the insertion and deletion of U residues, is catalyzed by sequential cleavage, U addition or removal, and ligation reactions and is directed by complementary guide RNAs. We have purified a approximately 20S enzymatic complex from Trypanosoma brucei mitochondria that catalyzes a complete editing reaction in vitro. This complex possesses all four activities predicted to catalyze RNA editing: gRNA-directed endonuclease, terminal uridylyl transferase, 3' U-specific exonuclease, and RNA ligase. However, it does not contain other putative editing complex components: gRNA-independent endonuclease, RNA helicase, endogenous gRNAs or pre-mRNAs, or a 25 kDa gRNA-binding protein. The complex is composed of eight major polypeptides, three of which represent RNA ligase. These findings identify polypeptides representing catalytic editing factors, reveal the nature of this approximately 20S editing complex, and suggest a new model of editosome assembly.

RNA editing in kinetoplastid protozoa

Mitochondrial transcripts in kinetoplastids undergo remarkable posttranscriptional editing by uridylate insertion and deletion. The often dramatic remodeling of pre-mRNA sequences is directed by small guide RNAs (gRNAs) to produce mature mRNAs. In vitro analyses of editing have been used to determine the mechanism of editing and show that editing occurs by a series of enzyme-catalyzed steps. They also show that chimeric gRNA/mRNA molecules are not editing intermediates as proposed but are aberrant end products of editing. The complexes and molecules that catalyze editing are now being identified and characterized. The origin of editing, its developmental regulation which helps control the switching between terminal respiratory systems during the life cycle of trypanosomes, and other areas for future study are discussed.

Guide RNA-independent and guide RNA-dependent uridine insertion into cytochrome b mRNA in a mitochondrial lysate from Leishmania tarentolae. Role of RNA secondary structure

A primer extension assay was used for the detection of uridine insertions occurring in vitro in synthetic pre-edited cytochrome b mRNA during incubation with a Leishmania tarentolae mitochondrial extract. Two different activities were detected that inserted uridines within the first two editing sites: one that is dependent on the secondary structure of the mRNA but is independent of both exogenous and endogenous guide RNA, and a second that does not put the same structural constraints on the mRNA, but is dependent on the presence of a cognate guide RNA.

Experimental verification of the secondary structures of guide RNA-pre-mRNA chimaeric molecules in Trypanosoma brucei

RNA editing in kinetoplastid organisms is an RNA-processing reaction that adds and deletes U nucleotides at specific sites in mitochondrial pre-mRNAs. The edited sequence is specified by guide RNAs and the processing presumably occurs within a high-molecular-mass ribonucleoprotein complex containing several enzymatic activities. Although the mechanism is not currently known, potential intermediates or by-products of the editing process are chimaeric RNAs where guide (g) RNAs are covalently attached, via their non-encoded U-tail, to their cognate pre-mRNAs. We determined the secondary structures of three different ATPase 6 chimaeras of Trypanosoma brucei using a set of structure-sensitive chemical and enzymatic probes. The experiments revealed a bipartite domain structure consisting of a gRNA/pre-mRNA interaction hairpin and an independently folding mRNA stem/loop in all three RNAs. The connecting U-tail was a determinant for the length of the interaction stems with the oligo(U) nucleotides base pairing to internal gRNA sequences. The probed structures have calculated delta G27o values of -92 kJ/ mol to -134 kJ/mol, somewhat less stable than the predicted minimal free energy structures and support previously proposed models for the interaction between gRNAs and pre-mRNAs. Optical melting studies indicated additional, higher order structural features for all three molecules with four defined melting transition between 10 degrees C and 90 degrees C. A comparison of CD spectra in the absence and presence of mitochondrial protein extracts demonstrated no gross structural changes of the RNA structures induced by the association with polypeptides.

Distinct differences in the requirements for ribonucleoprotein complex formation on differentially regulated pre-edited mRNAs in Trypanosoma brucei

Incubation of synthetic pre-edited mRNAs with extracts of Trypanosoma brucei mitochondria results in a family of specific, stable ribonucleoprotein (RNP) complexes that can be visualized by non-denaturing gel electrophoresis. We compared complexes that form with a substrate corresponding to the ATPase 6 (A6) pre-mRNA 3' region that is edited in both bloodstream and procyclic forms with a substrate corresponding to the 5' end of apocytochrome b (CYb) pre-mRNA that is edited only in procyclic (insect) forms. Four to five complexes are detected with both substrates and are specific since competition with homologous but not heterologous substrates prevents their formation. Formation of the CYb complex, however, is more sensitive to heterologous RNAs. In addition, the A6 substrate is more effective at preventing formation of CYb complexes than the converse. CYb complex formation is also more sensitive to divalent cation and salt concentrations and formation of one A6 component has a temperature optimum of 37 degrees C while that of CYb is 27 degrees C.

RNA editing: a mechanism for gRNA-specified uridylate insertion into precursor mRNA

In the mitochondria of trypanosomatid protozoa the precursors of messenger RNAs (pre-mRNAs) have their coding information remodeled by the site-specific insertion and deletion of uridylate (U) residues. Small trans-acting guide RNAs (gRNAs) supply the genetic information for this RNA editing. An in vitro system was developed to study the mechanism of U insertion into pre-mRNA. U-insertion editing occurs through a series of enzymatic steps that begin with gRNA-directed pre-mRNA cleavage. Inserted U's are derived from free uridine triphosphate and are added to the 3' terminus of a 5' pre-mRNA cleavage product. gRNA specifies edited RNA sequence at the subsequent ligation step by base pairing-mediated juxtaposition of the 3' cleavage product and the processed 5' cleavage product. gRNA/pre-mRNA chimeras, purported intermediates, seem to be abortive end products of the same reaction.

Trypanosome U-deletional RNA editing involves guide RNA-directed endonuclease cleavage, terminal U exonuclease, and RNA ligase activities

We have studied the mechanism of accurate in vitro RNA editing of Trypanosoma brucei ATPase 6 mRNA, using four mRNA-guide RNA (gRNA) pairs that specify deletion of 2, 3, or 4 U residues at editing site 1 and mitochondrial extract. This extract not only catalyzes deletion of the specified number of U residues but also exhibits a novel endonuclease activity that cleaves the input pre-mRNA in a gRNA-directed manner, precisely at the phosphodiester bond predicted in a simple enzymatic model of RNA editing. This cleavage site is inconsistent with a chimera-based editing mechanism. The U residues to be deleted, present at the 3' end of the upstream cleavage product, are then removed evidently by a 3' U-specific exonuclease and not by a reverse reaction of terminal U transferase. RNA ligase can then join the mRNA halves through their newly formed 5' P and 3' OH termini, generating mRNA faithfully edited at the first editing site. This resultant, partially edited mRNA can then undergo accurate, gRNA-directed cleavage at editing site 2, again precisely as predicted by the enzymatic editing model. All of these enzymatic activities cofractionate with the U-deletion activity and may reside in a single complex. The data imply that each round of editing is a four-step process, involving (i) gRNA-directed cleavage of the pre-mRNA at the bond immediately 5' of the region base paired to the gRNA, (ii) U deletion from or U addition to the 3' OH of the upstream mRNA half, (iii) ligation of the mRNA halves, and (iv) formation of additional base pairing between the correctly edited site and the gRNA that directs subsequent nuclease cleavage at the next editing site.

RNA editing: how a message is changed

Considerable progress has been made in unraveling the mechanistic features of RNA editing processes in a number of genetic systems. Recent highlights include the identification of the catalytic subunit of the mammalian apolipoprotein B mRNA editing enzyme as a zinc-dependent cytidine deaminase that binds to RNA, the demonstration that adenosines in brain glutamate receptor pre-mRNAs are converted into inosines and that double-stranded RNA A deaminase (dsRAD), the candidate enzyme, is another zinc-dependent RNA nucleotide deaminase, and a mounting body of evidence for a cleavage-ligation mechanism for U insertion/deletion editing in kinetoplastid protozoa.

Direct visualization of uridylate deletion in vitro suggests a mechanism for kinetoplastid RNA editing

Deletion of uridylates from the 3'-most editing site of synthetic ATPase 6 pre-mRNA can be visualized directly by coincubation of a radiolabeled substrate RNA and a synthetic gRNA in 20S fractions of T.brucie mitochondrial lysates. Substrate RNA cleavage is gRNA directed and occurs 3' to the uridylates to be deleted. U residues appear to be sequentially removed from the 3' end of the 5' cleavage product prior to religation of the two pre-mRNA halves. gRNA/mRNA chimeric molecules are also produced. Time course experiments indicate that chimeras appear after cleavage intermediates and edited product. Furthermore, a mutant gRNA promotes formation of edited product but not detectable chimeras. Our results suggest a model for kinetoplastid RNA editing in which chimeric molecules are nonproductive end products of editing and not intermediates that serve as a repository for deleted U's.

Trypanosoma brucei RNA editing. A full round of uridylate insertional editing in vitro mediated by endonuclease and RNA ligase

RNA editing in kinetoplastids is the post-transcriptional insertion and deletion of uridylate residues in mitochondrial transcripts, directed by base pairing with guide RNAs. Models for editing propose transesterification or endonuclease plus RNA ligase reactions and may involve a guide RNA-mRNA chimeric intermediate. We have assessed the feasibility of the enzymatic pathway involving chimeras in vitro. Cytochrome b chimeras generated with mitochondrial extract were first found to have junctions primarily at the major endonuclease cleavage sites, supporting the role of endonuclease in chimera formation. Such cytochrome b chimeras are then specifically cleaved by extract endonuclease within the oligo(U) tract at the editing site, and the mRNA cleavage products are then joined by RNA ligase to generate partially edited mRNAs with uridylate residues transferred to an editing site. These in vitro generated partially edited mRNAs mimic partially edited mRNAs generated in vivo. Specific endonuclease cleavage in the editing region of the partially edited RNA demonstrates the potential for further in vitro editing. Finally, sensitivity to various ATP analogs suggests that all editing-like activities reported thus far utilize a mechanism involving RNA ligase.

A possible role for the guide RNA U-tail as a specificity determinant in formation of guide RNA-messenger RNA chimeras in mitochondrial extracts of Crithidia fasciculata

Chimeric g(uide) RNA:pre-mRNA molecules are potential intermediates of the RNA editing process in kinetoplastid mitochondria. We have studied the characteristics of chimeric molecules formed in mitochondrial extracts of the insect trypanosomatid Crithidia fasciculata which had been supplied with synthetic NADH dehydrogenase (ND) subunit-7 gRNA and pre-mRNA variants. The ability of a gRNA to participate in chimera formation in this system depends on the possibility of base pairing with the pre-mRNA via the anchor sequence, but not on the presence of a U-tail or a full-length informational part. Chimeras formed with a specific gRNA:pre-mRNA pair displayed a large variation in length, due to variably sized 3' end truncations of the gRNA moieties and variation in the sites in the pre-mRNA to which the gRNAs were attached. Surprisingly, the presence of a U-tail in the gRNA for a large part determined the specificity of the linkage. In 60% of the cases gRNAs possessing a U-tail of at least one residue were attached to an editing site, whereas 75% of the gRNAs without Us were attached to non-editing sites. Furthermore, the chimera forming activity was greatly stimulated by the addition of ATP but not by AMP-CPP, an ATP-analogue with a non-hydrolyzable alpha-beta phosphate bond. This suggests the involvement in the chimera formation of an RNA ligase.

Guide RNA-mRNA chimeras, which are potential RNA editing intermediates, are formed by endonuclease and RNA ligase in a trypanosome mitochondrial extract

RNA editing in kinetoplast mitochondrial transcripts involves the insertion and/or deletion of uridine residues and is directed by guide RNAs (gRNAs). It is thought to occur through a chimeric intermediate in which the 3' oligo(U) tail of the gRNA is covalently joined to the 3' portion of the mRNA at the site being edited. Chimeras have been proposed to be formed by a transesterification reaction but could also be formed by the known mitochondrial site-specific nuclease and RNA ligase. To distinguish between these models, we studied chimera formation in vitro directed by a trypanosome mitochondrial extract. This reaction was found to occur in two steps. First, the mRNA is cleaved in the 3' portion of the editing domain, and then the 3' fragment derived from this cleavage is ligated to the gRNA. The isolated mRNA 3' cleavage product is a more efficient substrate for chimera formation than is the intact mRNA, inconsistent with a transesterification mechanism but supporting a nuclease-ligase mechanism. Also, when normal mRNA cleavage is inhibited by the presence of a phosphorothioate, normal chimera formation no longer occurs. Rather, this phosphorothioate induces both cleavage and chimera formation at a novel site within the editing domain. Finally, levels of chimera-forming activity correlate with levels of mitochondrial RNA ligase activity when reactions are conducted under conditions which inhibit the ligase, including the lack of ATP containing a cleavable alpha-beta bond. These data show that chimera formation in the mitochondrial extract occurs by a nuclease-ligase mechanism rather than by transesterification.

Trypanosoma brucei mitochondrial guide RNA-mRNA chimera-forming activity cofractionates with an editing-domain-specific endonuclease and RNA ligase and is mimicked by heterologous nuclease and RNA ligase

RNA editing in trypanosomes has been proposed to occur through transesterification or endonuclease cleavage and RNA ligation reactions. Both models involve a chimeric intermediate in which a guide RNA (gRNA) is joined through its 3' oligo(U) tail to an editing site of the corresponding mRNA. Velocity centrifugation of Trypanosoma brucei mitochondrial extracts had been reported to completely separate the gRNA-mRNA chimera-forming activity from endonuclease activity (V. W. Pollard, M. E. Harris, and S. L. Hajduk, EMBO J. 11:4429-4438, 1992), appearing to rule out the endonuclease-RNA ligase mechanism. However, we show that an editing-domain-specific endonuclease activity does cosediment with the chimera-forming activity, as does the RNA ligase activity, but detection of the specific endonuclease requires reducing assay conditions. This report further demonstrates that the T. brucei chimera-forming activity is mimicked by mung bean nuclease and T4 RNA ligase. Using cytochrome b (CYb) preedited mRNA and a model CYb gRNA, we found that these heterologous enzymes specifically generate CYb gRNA-mRNA chimeras analogous to those formed in the mitochondrial extract. These combined results provide support for the endonuclease-RNA ligase mechanism of chimera formation.