Trans -splicing of a mutated glycosylasparaginase mRNA sequence by a group I ribozyme deficient in hydrolysis Eirik W. Lundblad 1 , Peik Haugen 1 and Steinar D. Johansen 1,2 1 Department of Molecular Biotechnology, RNA Research group, Institute of Medical Biology, University of Tromsø, Norway; 2 Department of Fisheries and Natural Sciences, Bodø Regional University, Norway RNA reprogramming represents a new concept in correcting genetic defects at the RNA level. However, for the technique to be useful for therapy, the level of reprogramming must be appropriate. To i mprove the efficiency of g roup I ribozyme- mediated RNA reprogramming, when using the Tetrahy- mena ribozyme, regions complementary to the target RNA have previously been extended in length and accessible sites in the target RNAs have been identified. As an alternative to the Tetrahymena model ribozyme, the D iGIR2 group I ribozyme, derived from a mo bile group I intron i n rDNA of the myxomycete Didymium iridis , represents a new and attractive tool in RNA reprogramming. We r eported recently that the deletion of a structural element within the P9 domain of DiGIR2 turns off hydrolysis at the 3¢ splice site (side reaction) without affecting self-splicing [Haugen, P., Andreassen, M., Birgisdottir, A ˚ .B.&Johansen,S.D.(2004) Eur. J. Biochem. 271 , 1015–1024]. Here we a nalyze the potential of the modified ribozyme, deficient in hydrolysis at the 3¢ splice site, for applicatio n in group I ribozyme-medi- ated trans-splicing of RNA. The improved ribozyme cata- lyses both cis-splicing and trans-splicing in vitro of a human glycosylasparaginase mRNA sequence with the same effi- ciency as the original DiGIR2 ribozyme, but without detectable levels of the unwanted hydrolysis. Keywords: glycosylasparaginase mRNA; group I i ntron; ribozyme hydrolysis; RNA reprogramming; trans-splicing. Group I ribozyme-mediated RNA r eprogramming by trans-splicing, has been successfully carried out using the Tetrahymena ribozyme and various target RNAs [1–5]. The trans-splicing reaction is similar to the self-splicing reaction normally catalysed by group I introns [6], except that the 5¢ exon is presented in trans and a corrected 3¢ exon is attached to the r ibozyme. Ligation o f these exons produces the chimerical transcript that can be translated into a functional protein. RNA r eprogramming is guided by a region complementary to the target RNA (internal guide sequence, IGS) located within the ribozyme, and the specificity and efficiency of trans-splicing have mainly been improved by extending the IGS [2,7–9]. In addition, group I ribozymes with randomised IGSs are used to identify regions on the target RNAs that are accessible [1,4,10–13]. I n spite of recent advances and significant efforts to optimize trans- splicing reactions, the RNA reprogramming in cells remains inefficient. Moreover, group I ribozymes, including the Tetrahymena ribozyme, catalyze additional reactions that directly c ompete with sp licing and prob ably lower the efficiency of trans-splicing. Most pronounced is the 3¢ splice site hydrolysis of precursor RNAs [14–16], which is catalysed by the Tetrahymena ribozyme at a relatively high rate [16,17]. Hydrolysis results i n the formation of full- length intron RNA circles, which are commonly detected both in vitro and in vivo in a number of group I introns [17– 19]. Designing ribozymes that catalyse little or no competing side reactions can therefore prove valuable in the search for better ribozyme tools that can be used in RNA reprogram- ming. DiGIR2 i s the splicing ribozyme derived from the twin-ribozyme g roup I intron D ir.S956-1 i n Didymium ribosomal DNA (Fig. 1A) [20,21]. DiGIR2 represents the group IE intron subgroup with clear distinction in structure c ompared t o t he distantly related Tetrahymena group IC1 intron [16,18,19]. We recently reported t hat deletion of the P9.2 paired element in the DiGIR2 ribozyme (Fig. 1B) s ignificantly reduces hydrolytic clea- vage at the 3¢ splice site without affecting the self-splicing activity in cis-splicing constructs [16]. The remarkable loss of unwanted side reactions, apparently without compromising splicing, identifies the new ribozyme con- struct (denoted DiGIR2DP9.2) as a potential improved tool in group I ribozyme-mediated trans-splicing of RNA. Here we set out to investigate t he ability and efficiency of DiGIR2 a nd DiGIR2DP9.2 to trans-splice RNA molecules. Trans-splicing ribozymes were construc- ted and targeted against a mutated glycosylasparaginase Correspondence to S. Johansen, Department of Molecular Biotech- nology, Institute of Medical Biology, University of Tromsø, 9037 Tromsø, Norway. Fax: + 47 77 64 53 50, Tel.: + 47 77 64 53 67, E-mail: Steinar.Johansen@fagmed.uit.no Abbreviations: AGU, aspartylglycosaminuria; EGS, extended guide sequence; GA, glycosylasparaginase; IGS, internal guide sequence; nt, nucleotide; RPA, ribonuclease protection analysis. Note: The oligonucleotide sequences used in this work are available on request and as a supplement at the RNA Research Group’s website at http://www.fagmed.uit.no/info/imb/amb (Received 15 April 2004, revised 9 August 2004, accepted 25 October 2004) Eur. J. Biochem. 271, 4932–4938 (2004) Ó FEBS 2004 doi:10.1111/j.1432-1033.2004.04462.x (GA) mRNA sequence. Mutations in GA cause the human lysosomal storage disease aspartylglycosaminuria (AGU) [22]. Experimental procedures Plasmid constructions and in vitro mutagenesis The cis -splicing construct p DiGIR2 AGU was made by combining two different PCR products. The first product contains the T7 promoter, 15 nucleotides (nt) from the human GA open reading frame (ORF) and the DiGIR2 splicing ribozyme, and was generated from the pDiGIR2 template [20] using the primer combination OP340/341. The second product (108 nt) was amplified from a cloned human GA cDNA template using the primer combination OP342/346. The two PCR products were blu nted, phos- phorylated and ligated using the Sure Clone Ligation Kit (Amersham Biosciences, Piscataway, NJ, USA). Fin ally, a new PCR product w as generated f rom the ligation mix using the oligo primers OP341/346 and subsequently cloned into pUC18. The P9.2 hairpin was deleted from the pDiGIR2 AGU by using the Quick Change site-directed mutagenesis kit (Stratagene, Cedar Creek, TX, USA) and OP296/297, generating pDiGIR2DP9.2 AGU. The trans- splicing constructs were made by PCR amplification with Pfu DNA polymerase (Promega, Madison, WI, USA) using pDiGIR2 AGU and pDiGIR2DP9.2 AGU as templates, generating the trans-splicing plasmid versions of pDiGIR2 AGU and pDiGIR2DP9.2 AGU, respectively. Here, the primer combinations OP1191/1192 and OP1191/ 1202 were used. The forward primer w as designed with two sequences complementary to the target RNA [8 nt IGS and 9 nt extended guide sequence (EGS)] separated by a 5 nt wobble region. The reverse primers were designed with alternative codons, of which the first 5 nt in the 3¢ exon are able to form a P10 helix with nucleotides in the IGS-wobble region. The PCR products were digested with NotIand BamHI, gel extracted (QIAquick gel extraction kit; QIAGEN, Gmbh, Germany), and ligated downstream of Fig. 1. Constructs and structural features of the DiGIR2 ribozyme. (A) Organization of the twin-ribozyme intron (Dir.S956-1) into group I ribozyme motifs (DiGIR1 and DiGIR2) and the I-DirI homing endonuclease gene, as w ell as the two versions of the DiGIR2 ribozyme used in th is study. The 5¢ and 3¢ splice sites (SS) are indicated, and flanking exon sequences are shown as open boxes. (B) Secondary structure of the DiGIR2 ribozyme [16]. Boxed nucleotides in P9.2 are deletedinDiGIR2DP9.2. Intron RNA nucle- otides and exon nucleotides are presented as uppercase and lowercase letters, respectively. Ó FEBS 2004 Hydrolysis deficient trans-splicing group I ribozyme (Eur. J. Biochem. 271) 4933 the CMV- and T7-promoters into corresponding sites in a pDNA3.1(–) vector (Invitrogen, Norway), which had the nucleotide s equences between the NheIandXbaIsites deleted to bring the inserts closer to the T7 RNA polymerase t ranscription initiation site. The target GA RNA, containing the prevalent Finnish mutation (Fig. 2 A), was PCR amplified with Pfu ultra HF DNA polymerase (Stratagene) using the primer combination OP1219/1220 (containing NheIandBamHI sites, respectively). The PCR product was digested with NheIandBamHI, gel extracted (Qiagen gel extraction kit), a nd ligated downstream o f t he CMV- and T7-promoters into corresponding sites i n the pDNA3.1(–) vector (Invitrogen). All constructs were con- firmed by automatic sequencing by the ABI PRISM BigDyeTerminator Cycle Sequencing Ready Reaction Kit (PerkinElmer, Norway) running on an ABI Prism 377 system (PerkinElmer). Oligonucleotide sequences used in this work are available on request or as a supplement at the RNA Research Group’s website at http://www.fagmed. uit.no/info/imb/amb. In vitro transcription, splicing reactions and RT-PCR analysis PrecursorRNAsforcis-splicing analyses were transcribed from T7 promoters off BamHI-linearized pDiGIR2, pDiGIR2 AGU and pDiGIR2DP9.2 AGU plasmids. [ 35 S]CTP[aS] (10 lCiÆlL )1 ; Amersham Biosciences) was uniformly incorpora ted into the RNA transcripts. R NA splicing was performed under self-splicing conditions essentially as described [20]. Samples were separated on 8 M urea/5% polyacrylamide gels, followed by autoradio- graphy. To obtain the sufficient amounts o f RNA the constructs were transcribed at 8 m M MgCl 2 , resulting in some splicing activity a t time point 0. To analyse the ligated exon sequences from pDiGIR2 AGU and pDi- GIR2DP9.2 AGU RNAs, RNA corresponding to ligated exons was g el isolated and e luted in 400 lLofelution buffer (0.3 M NH 4 Ac, 0.1% SDS, 10 m M Tris pH 8 and 2.5 m M EDTA pH 8) overnight on a rotary mixer at 4 °C. RNA w as subsequently filtered through a 0.45 l M A B Fig. 2. Cis-splicing experiments of DiGIR2- derived ribozymes inserted into GA RNA sequences. ( A) Top; schematic map of the human GA ORF indicating the intron inser- tion site at position 436. Mutations at posi- tions 488, 800 and 916 are frequently associated with the most common lysosomal degradation disorder AGU found in the Fin- nish, Spanish/American and American popu- lations, respectively [22]. Middle; sche matic drawing of RNA transcripts generated from constructs containing the DiGIR2 or DiGIR2DP9.2 ribozymes. The ribozyme internal guide sequence (IGS) sequence was adapted to the heterologous exon sequence. Bottom; similarities between flanking 5¢ and 3¢ exon sequences are noted between the Didymium rDNA and human GA ORF. Underlined positions are identical. (B) Left; time course cis-splicing experiment (0–30 min) of DiGIR2 [in small subunit (SSU) rRNA] and the two GA ORF intron constructs DiGIR2 AGU and DiGIR2DP9.2 AGU. The RNA species is present at time point 0 due to some splicing activity during transcription (Experimental procedures). Right; represen- tative result of a ligated exon sequence ladder obtained from an RT-PCR analysis of RNA 5. The DNA sequence is similar to the RNA sequence shown below. The ligated exon junction is marked by an arrowhead. 4934 E. W. Lundblad et al.(Eur. J. Biochem. 271) Ó FEBS 2004 single use filter (Millipore, Ireland), ethanol precipitated and subjected to reverse transcription using t he First Strand Synthesis kit (Amersham Biosciences) and OP346. Ligated exons (120 bp) were amplified with OP346/421, separated on a high percentage agarose gel, eluted using the Agarose Gel Extraction kit (Boehringer Mannheim, Mannheim, Germany), and finally cloned into pUC18. Two independent ligated exon cDNA clones from each of the pDiGIR2 AGU and pDiGIR2DP9.2 AGU were manually sequenced using the Thermo Sequenase sequen- cing kit (Amersham Biosciences) and [ 33 P]ddNTPs (GATC; 450 lCiÆmL )1 ) as the label. Precursor RNAs for trans-splicing analyses (Fig. 3) were in vitro tran- scribed from T7 promoters off Bam HI-linearized plas- mids without [ 35 S]CTP l abeling. The t arget G A RNA transcript was  1050 n t. A similar RT-PCR experiment, as described a bove, was performed on the t rans-spliced products, but using the primer OP1194 in the RT reaction and OP1193/1194 in amplification. Trans-splicing and ribonuclease protection analyses In trans-splicing experiments, unn labeled DiGIR2 AGU or DiGIR2DP9.2 AGU RNAs and PAGE-purified GA RNA were mixed in a 3 : 1 ratio. Two microliters of 5· low-salt buffer ( 40 m M Tris/HCl pH 7.5 , 200 m M KCl, 2m M spermidine, 5 m M dithiothreitol, 10 m M MgCl 2 , 0,2 m M GTP) was added and the volume was adjusted to 10 lLwithwater.Thetrans-splicing mix was incuba- ted at 3 7 °C for 3 h. Ribonuclease protection a nalysis (RPA) was performed on 5 lLoftrans-splicing RNA- mix by the RNase p rotection kit (Roche Applie d Science, Penzberg, Germany) according to the manufacturer’s instructions. The RPA probe was generated from the RT-PCR product of in vitro trans-spliced GA RNA (see above) cloned i nto the pGEM-T easy vector (Promega). This plasmid was linearized and transcribed from the SP6 promoter, labelling with [ 35 S]CTP as described above, to get a RPA probe of larger size than the probe fragment protected by trans-spliced RNA in analysis by RPA. The transcribed RPA probe was  500 nt (Fig. 4B). RPA samples were separated on 8 M urea/5% polyacrylamide gels, followed by autoradio- graphy (Fig. 3B) and phosphoimager quantitation (Fuji BAS 5000 system; IMAGE GAUGE 4.0 software). The cytosine content in the part of the RPA probe protected by the different sized RNAs was calculated and included as a theoretical value to make the intensities of different sized bands comparable. The amount of reprogrammed product (RNA 2) was calculated as a fraction (in A B Fig. 3. Design of trans-splicing ribozyme c on- struct s. (A) The ribozyme contains the internal guide sequence (IGS) and extended guide sequence (EGS), which base-pairs to the complementary sequence in GA mRNA upstream of the mutation. The ribozyme catalyzes the coupled cleavage of m utated mRNA and the ligation of the restorative 3¢ exon to the remaining 5¢ exon. (B) The ribo- zyme constructs used contain silent mutations (underlined) introduced by alternative codons, andasequencetagusedinRT-PCRdetection (boxed nucleotides) [4]. Ó FEBS 2004 Hydrolysis deficient trans-splicing group I ribozyme (Eur. J. Biochem. 271) 4935 percentage) of reprogrammed product (RNA 2) + target GA RNA (RNA 3). The amount of trans-splicing ribozymes that had undergone the r eaction [ calculated as the amount o f reprogrammed product (RNA 2) as a fraction (in percentage) of reprogrammed product (RNA 2) + tra ns-splicing ribozyme (RNA 4)] was similar fo r the t wo different ribozyme c onstructs (data not sh own). Three parallels of RPA experiments were performed. AB Fig. 4. Trans-splicing experiments of DiGIR2-derived ribozymes including the GA RNA as target sequence. RT-PCR and ribonu clease protection analysis (RPA) of trans-spliced GA mRNA generated by DiGIR2 AGU and DiGIR2DP9.2 AGU ribozymes. (A) Top; RT-PCR products from in vitro trans-splicing analyses with DiGIR2 AGU and D iGIR2DP9.2 AGU corresponding to reprogrammed products of the expected size (361 nt). The negative control (Neg. Ctrl.) contains fi rst-strand synthesis master mix. Below; representative re sult of trans-spliced GA RNA sequence obtained from RT-PCR. The trans-splicing junction is marked by an arrow. (B) Top; representative result of major RNA-species (numbered 1–4) detected in RPA. RNA 1, undigested probe; RNA 2, trans-spliced GA mRNA; RNA 3, major GA band; RNA 4, major DiGIR2 AGU and DiGIR2DP9.2 AGU band. Additional bands result from degradation of RNA during incubation at trans-splicing and RPA hybridization conditions. Below; quantitation of RPA of trans-spliced GA mRNA generatedbyDiGIR2AGUandDiGIR2DP9.2 AGU. Comparative quantitative data were collected from three indepen dent assays. The trans -splicing efficie ncy (perce ntage) was c alculated b y dividing the trans- splicing band (RNA 2) ·100bythesumofthetrans-splicing band (RNA 2) and the major GA band (RNA 3). As the amount of DiGIR2 AGU and DiGIR2DP9.2 AGU r iboz ymes ad ded in t he trans -splicing reactions was a pproximately identical, the a ddition of t he ribozyme b and (RNA 4 ) in t he fractions denominator gave approximately identical comparative results (data not shown). 4936 E. W. Lundblad et al.(Eur. J. Biochem. 271) Ó FEBS 2004 Results and Discussion In vitro self-splicing of DiGIR2 and DiGIR2DP9.2 ribozymes from heterologous transcripts To test the potential of DiGIR2 and DiGIR2DP9.2 in RNA reprogramming, the ribozymes were first inserted in cis into heterologous exons that represent therapeutic, relevant RNA sequences. The ribozymes were inserted between positions 436 and 437 of the human GA ORF sequence (Fig. 2A) and tested for splicing a ctivity in vitro. The most common disorder of glycoprotein degradation, AGU, is caused by mutations in GA [22]. The nucleotides flanking the intron insertion site in GA RNA, upstream of the prevalent Finnish AGU mutation, is similar to the nucleotides flanking the wild type Didymium rDNA insertion site (Fig. 2A). In the corresponding splicing constructs, t he IGS was modified from GGCCGCfiGGUCUU in order to adapt the ribo- zymes to the GA 5¢ exon. Figure 2B shows that the IGS- modified DiGIR2 ribozyme excised itself from the precursor RNA, and in the same process correctly ligated the surrounding exons (DiGIR2 AGU, Fig. 2B). Bands that represent intron circle (RNA 1), precursor RNA (RNA 2), 5¢ exon–intron (RNA 3), f ree intron (RNA 4), ligated exons (RNA 5), and free 3¢ exons (RNA 7), are visible. The small 5¢ exon (RNA 6) was run off the gel. The IGS-modified DiGIR2DP9.2 AGU transcript generated a band pattern analogous to DiGIR2 AGU, except for the hydrolysis- dependent RNA species (RNAs 1, 3 and 7) that were absent in the reaction (Fig. 2B). In conclusion, these results show that both the DiGIR2 and DiGIR2DP 9.2 ribozymes accu- rately self-splice when inserted into foreign exons in cis. In vitro trans -splicing of mutated GA mRNA sequences using DiGIR2 and DiGIR2DP9.2 ribozymes To test whether the DiGIR2 and DiGIR2DP9.2 ribozymes are able to splice foreign exons also in trans, we targeted the ribozymes to position 436 (uracil) in the mutated GA mRNA (same site as in the cis-splicing experiment; Fig. 3) located upstream of the most common AGU mu tations (Fig. 2). The ribozymes were designed with modifications known to increase trans-splicing efficiency and specificity [1,4,9]; an IGS of 8 nt was used, and based on work by Sullenger and coworkers [5], a 9 nt EGS complementary to the GA mRNA target was added (Fig. 3) to further increase specificity and efficiency. Furthermore, a P10 helix of 5 nt were included as this i s shown to substantially increase trans-splicing efficiency of the Tetrahymena riboz yme [9]. Finally, the 3¢ exon that contains the c orrected AGU sequence was degenerated by alternative codons (Fig. 3) in order t o avoid strong inter molecular base-pairing to the region complementary to the target RNA [4]. The trans-splicing ribozymes and RNA targets were mixed and subjected to conditions that favour splicing (see above). In a n RT-PCR a pproach the trans-ligated exon products were amplified and DNA sequen ced to verify correct splicing (Fig. 4A). In order to quantify the amount of trans-spliced product, and compare the trans-splicing efficiency between DiGIR2 AGU and DiGIR2DP9.2 AGU, we performed analysis by RPA. The RPA probe was designed to hybridize to a 312 n t region located upstream of U436 in mutated GA mRNA and to a 49 nt region of the 3¢ exon in the ribozymes, resulting in a 361 nt protected region for the trans-spliced RNA. The probe was in vitro transcribed containing additional vector sequences in order to easily separate full-length probe from RNA fragments protected in analysis by RPA. Gel analyses of RPA products (Fig. 4B) confirmed the RT-PCR based experi- ment presented above of in vitro trans-splicing. The amount of trans-spliced products were ap proximately similar for DiGIR2AGU and DiGIR2DP9.2 AGU (1.8% and 2.0%, respectively). In summary, t he DiGIR2DP9.2 ribozyme deficient i n hydrolysis is able to perform trans-splicing with high fidelity in vitro at comparable rate compared to the wild-type derived DiGIR2 ribozyme. The former ribozyme is smaller in size and lacks the hydrolytic processing known to compete with intro n splicing [18]. Previous works on RNA reprogramming have focused on using the Tetrahymena ribozyme as t he tool. Our findings demonstrate that t he DiGIR2 ribozyme (and its derivatives), in which 3¢ splice site hydrolysis can be assigned to defined structures within the intron [18], represent an interesting alternative to the Tetrahymena ribozyme. Although the 3¢ splice site hydro- lysis side reaction is under control we realize that the challenge for the future will be to increase the fraction of reprogrammed mRNA. Here, experiments that involve selection for better target accessibility a nd re programming [4,13,23] will be crucial. Acknowledgements This work was f unded by grants f rom The Norwegian Research Council, The Norwegian Cancer Society, The Aakre Foundation for Cancer Research, and Simon Fougner Hartmanns Foundation. We thank Dr Ole K. Tollersrud and Dr Hilde Monica Frostad Riise for the glycosylasparaginase cDNA plasmids. References 1. Watanabe, T. & Sullenger, B.A. (2000) RNA repair: a novel approach to gene therapy. Adv. Drug. Delivery Rev. 44, 109–118. 2. Rogers, C.S., Vanoye, C.G., S ullenger, B.A. & George, A.L. (2002) Functional repair of a mutant chloride channel using a trans-splicing ribozyme. J. Clin. Invest. 110, 1783–1789. 3. Long, M.B., Jones, J.P., Sullenger, B.A. & Byun, J. (2003) Ribo- zyme-mediated re vision o f RNA and DNA. J. Clin. In vest. 112, 312–318. 4. Einvik,C.,Fiskaa,T.,Lundblad,E.W.&Johansen,S.(2004) Optimization and a pplicat ion of the group I ribozym e trans-spli- cing reaction. Methods Mol. Biol. 252, 359–372. 5. Byun, J., Lan, N., Long, M. & Sullenger, B.A. (2003) Efficient and specific repair of s ickle beta-globin RNA by trans- splicing ribo- zymes. RNA 9, 1254–1263. 6. Cech, T.R. (1990) Self-splicing of group I introns. Annu. Rev. Biochem. 59, 543–568. 7. Zarrinkar, P.P. & Sullenger, B.A. (1999) Optimizing the substrate specificity of a group I intron ribozyme. Biochemistry 38, 3426– 3432. 8. Ayre,B.G.,Kohler,U.,Goodman,H.M.&Haseloff,J.(1999) Design of highly specific cytotoxins b y using trans-splicing ribo- zymes. Proc. Natl Acad. Sci. USA 96, 3507–3512. 9. Kohler,U.,Ayre,B.G.,Goodman,H.M.&Haseloff,J.(1999) Trans-splicing ribozymes for targeted gene delivery. J. Mol. Biol. 285, 1935–1950. Ó FEBS 2004 Hydrolysis deficient trans-splicing group I ribozyme (Eur. J. Biochem. 271) 4937 10. Sullenger, B.A. (1995) Colocalizing r ibozymes with substrate RNAs to increase their efficacy as gene inhibitors. Appl. Biochem. Biotechnol. 54, 57–61. 11. Lan, N., Howrey, R.P., Lee, S.W., Smith, C.A. & Sullenger, B.A. (1998) Ribozyme-mediated repair of sickle beta-globin mRNA’s in erythrocyte precursors. Science 280, 1593–1596. 12. Lan,N.,Rooney,B.L.,Lee,S.W.,Howrey,R.P.,Smith,C.A.& Sullenger, B.A. ( 2000) Enhancing RNA repair efficiency by com- bining trans-splicing ribozymes that recognize different accessible sites on a target RNA. Mol. Ther. 2, 245–255. 13. Ayre, B.G., Kohler, U., Turgeon, R. & Haseloff, J. (2002) Opti- mization of trans-splicing ribozyme efficiency a nd specificity by in vivo genetic selection. Nucleic Acids Res. 30,e141,1–9. 14. Golden, B.L. & Cech, T.R. (1996) Conformational switches involved in orchestrating the successive steps of group I RNA splicing. Biochemistry 35, 3754–3763. 15. Zarrinkar, P.P. & Sullenger, B.A. (1998) Probing the interplay between the two steps of group I intron splicing: competition of exogenous guanosine with xG. Biochemistry 37, 18056–18063. 16. Haugen, P., An dreassen, M., Birgisdottir, A ˚ .B. & Johansen, S. (2004) Hydrolytic cleavage by a group I intron ribozym e is dependent on RNA structures not important for splicing. Eur. J. Biochem. 271, 1015–1024. 17. Nielsen, H., Fiskaa, T., Birgisdottir, A ˚ .B., Haugen, P., Einvik, C. & Johansen, S. (2003) The ability to form full-length intron RNA circles is a general property of nuclear group I intro ns. RNA 9, 1464–1475. 18. Haugen, P., De Jonckheere, J.F. & Johansen, S. (2002) Char- acterization of the self-splicing products of two complex Naegleria LSU rDNA group I introns containing homing endonuclease genes. Eur. J. Biochem. 269, 1641–1649. 19. Lundb lad, E.W., Einvik, C., Rønning, S., Haugli, K. & Johansen, S. (2004) Twelve group I introns in the same pre-rRNA transcript of the myxomycete Fuligo septica: RNA processing and evolution. Mol. Biol. Evol. 21 , 1283–1293. 20. Decatur, W.A., Einvik, C., Johansen, S. & Vogt, V.M. (1995) Two group I ribozymes with different func tions in a nu clear rDNA intron. EMBO J. 14, 4558–4568. 21. Einvik, C., Elde, M. & Johansen, S. (1998) G roup I twintrons: genetic elements in myxomycete and schizopyrenid amoebo- flagellate ribosomal DNAs. J. Biotechnol. 17 , 63–74. 22. Mononen,I.,Fisher,K.J.,Kaartinen,V.&Aronson,N.N.Jr (1993) Aspartylglycosaminuria: protein chemistry and mole cular biology o f the most common lysosomal storage disorder of gly- coprotein degradation. FASEB J. 7, 1247–1256. 23. Jones, J.T. & Sullenger, B.A. (1997) Evaluating and e nhancing ribozyme reaction efficiency in mammalian cells. Nat. Biotechnol. 15, 902–905. 4938 E. W. Lundblad et al.(Eur. J. Biochem. 271) Ó FEBS 2004 . Trans -splicing of a mutated glycosylasparaginase mRNA sequence by a group I ribozyme deficient in hydrolysis Eirik W. Lundblad 1 , Peik Haugen 1 and Steinar. splice site, for applicatio n in group I ribozyme- medi- ated trans-splicing of RNA. The improved ribozyme cata- lyses both cis-splicing and trans-splicing