An mRNA 3′ Processing Site Targets Downstream Sequences for Rapid Degradation in Chlamydomonas Chloroplasts
2002; Elsevier BV; Volume: 277; Issue: 5 Linguagem: Inglês
10.1074/jbc.m108979200
ISSN1083-351X
AutoresAmanda J. Hicks, Robert G. Drager, David C. Higgs, David B. Stern,
Tópico(s)Protist diversity and phylogeny
ResumoIn Chlamydomonas chloroplasts,atpB pre-mRNA matures through a two-step process. Initially, endonuclease cleavage occurs 8–10 nt downstream of the mature 3′ end, which itself lies at the end of a stem-loop-forming inverted repeat (IR) sequence. This intermediate product is then trimmed by a 3′ → 5′ exonuclease activity. Although the initial endonucleolytic cleavage by definition generates two products, the downstream product of atpB pre-mRNA endonucleolytic processing cannot be detected, even transiently. This product thus appears to be highly unstable, and it can be hypothesized that specific mechanisms exist to prevent its accumulation. In experiments described here, the atpB 3′ maturation site was placed upstream of reporter genes in vivo. Constructs containing both the IR and endonuclease cleavage site (ECS) did not accumulate the reporter gene mRNA, whereas constructs containing only the IR did accumulate the reporter mRNA. The ECS alone gave an intermediate result, suggesting that the IR and ECS act synergistically. Additional secondary structures were used to test whether 5′ → 3′ and/or 3′ → 5′ exonuclease activities mediated degradation. Because these structures did not prevent degradation, rapid endonucleolytic cleavages most likely trigger RNA destruction after ECS cleavage. On the other hand, fragments resulting from cleavage within the endogenousatpB mRNA could occasionally be detected as antisense transcripts of the adjacent reporter genes. Because endonuclease cleavages are also involved in the 5′ maturation of chloroplast mRNAs, where only the downstream cleavage product accumulates, it appears that chloroplast endoribonuclease activities have evolved mechanisms to selectively stabilize different ECS products. In Chlamydomonas chloroplasts,atpB pre-mRNA matures through a two-step process. Initially, endonuclease cleavage occurs 8–10 nt downstream of the mature 3′ end, which itself lies at the end of a stem-loop-forming inverted repeat (IR) sequence. This intermediate product is then trimmed by a 3′ → 5′ exonuclease activity. Although the initial endonucleolytic cleavage by definition generates two products, the downstream product of atpB pre-mRNA endonucleolytic processing cannot be detected, even transiently. This product thus appears to be highly unstable, and it can be hypothesized that specific mechanisms exist to prevent its accumulation. In experiments described here, the atpB 3′ maturation site was placed upstream of reporter genes in vivo. Constructs containing both the IR and endonuclease cleavage site (ECS) did not accumulate the reporter gene mRNA, whereas constructs containing only the IR did accumulate the reporter mRNA. The ECS alone gave an intermediate result, suggesting that the IR and ECS act synergistically. Additional secondary structures were used to test whether 5′ → 3′ and/or 3′ → 5′ exonuclease activities mediated degradation. Because these structures did not prevent degradation, rapid endonucleolytic cleavages most likely trigger RNA destruction after ECS cleavage. On the other hand, fragments resulting from cleavage within the endogenousatpB mRNA could occasionally be detected as antisense transcripts of the adjacent reporter genes. Because endonuclease cleavages are also involved in the 5′ maturation of chloroplast mRNAs, where only the downstream cleavage product accumulates, it appears that chloroplast endoribonuclease activities have evolved mechanisms to selectively stabilize different ECS products. inverted repeat endonuclease cleavage site untranslated receptor nucleotide(s) wild type Maturation of mRNA can involve multiple steps in both prokaryotic and eukaryotic systems, and results in functional transcripts with a stability and subcellular localization appropriate to their functions. RNA sequence and secondary structure, along with ribonucleases and a variety of accessory factors, are used to achieve these goals. Built into this process is the necessity to recognize nonfunctional RNAs and eliminate them; destruction of nonsense codon-containing mRNAs is a good example of the complexity of such surveillance mechanisms (reviewed in Ref. 1Czaplinski K. Ruiz-Echevarria M.J. Gonzalez C.I. Peltz S.W. Bioessays. 1999; 21: 685-696Crossref PubMed Scopus (92) Google Scholar). Our laboratory has focused on 5′ end and 3′ end maturation of chloroplast mRNAs, using both vascular plants and the unicellular green alga Chlamydomonas reinhardtii as models. In this respect, chloroplast transcripts have mostly prokaryotic features such as the lack of a trimethylguanosine 5′ cap, a 3′ stem-loop-forming inverted repeat (IR)1structure, and they are destabilized by polyadenylation (reviewed in Refs. 2Monde R.A. Schuster G. Stern D.B. Biochimie (Paris). 2000; 82: 573-582Crossref PubMed Scopus (113) Google Scholar and 3Schuster G. Lisitsky I. Klaff P. Plant Physiol. 1999; 120: 937-944Crossref PubMed Scopus (69) Google Scholar). Processing of chloroplast mRNAs, with rare exceptions, depends on nucleus-encoded proteins, several of which have been identified genetically through screens for plants orChlamydomonas strains unable to carry out photosynthesis. The phenotypes suggest that defects in intercistronic processing of polycistronic transcripts (4Felder S. Meierhoff K. Sane A.P. Meurer J. Driemel C. Plucken H. Klaff P. Stein B. Bechtold N. Westhoff P. Plant Cell. 2001; 13: 2127-2141Crossref PubMed Scopus (94) Google Scholar, 5Barkan A. Walker M. Nolasco M. Johnson D. EMBO J. 1994; 13: 3170-3181Crossref PubMed Scopus (203) Google Scholar) or in correct 5′ end maturation (6Vaistij F.E. Boudreau E. Lemaire S.D. Goldschmidt-Clermont M. Rochaix J.D. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 14813-14818Crossref PubMed Scopus (107) Google Scholar, 7Boudreau E. Nickelsen J. Lemaire S.D. Ossenbuhl F. Rochaix J.-D. EMBO J. 2000; 19: 3366-3376Crossref PubMed Scopus (132) Google Scholar, 8Drager R.G. Girard-Bascou J. Choquet Y. Kindle K.L. Stern D.B. Plant J. 1998; 13: 85-96Crossref PubMed Scopus (94) Google Scholar) can cause RNA instability and/or translational defects. Also, we have previously shown that correctly 3′ end processed mRNA is translated preferentially in Chlamydomonas chloroplasts (9Rott R. Levy H. Drager R.G. Stern D.B. Schuster G. Mol. Cell. Biol. 1998; 18: 4605-4611Crossref PubMed Scopus (39) Google Scholar). Therefore, 5′ and 3′ end maturation are essential steps in chloroplast gene expression. Biogenesis of the Chlamydomonas chloroplast atpBmRNA is particularly well characterized. The atpB gene contains a somewhat unusual promoter, with essential elements included in the 5′ UTR (10De Klein U. Camp J.D. Bogorad L. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 3453-3457Crossref PubMed Scopus (61) Google Scholar). Following transcription, two 5′ ends can be found in the mature mRNA (11Blowers A.D. Ellmore G.S. Klein U. Bogorad L. Plant Cell. 1990; 2: 1059-1070PubMed Google Scholar), a situation also found for many other chloroplast transcripts and presumed or proven to result from a primary transcript undergoing partial endonucleolytic cleavage. The region spanning from 10–89 nt downstream of the atpB stop codon contains an IR sequence predicted to form an AU-rich stem-loop structure, and the mature 3′ end is found 3–4 nt downstream of the IR (12Stern D.B. Radwanski E.R. Kindle K.L. Plant Cell. 1991; 3: 285-297PubMed Google Scholar, 13Stern D.B. Kindle K.L. Mol. Cell. Biol. 1993; 13: 2277-2285Crossref PubMed Scopus (60) Google Scholar). Deletions that destabilize this structure cause RNA instability (12Stern D.B. Radwanski E.R. Kindle K.L. Plant Cell. 1991; 3: 285-297PubMed Google Scholar); however the sequence can be functionally replaced by an IR from spinach chloroplasts (13Stern D.B. Kindle K.L. Mol. Cell. Biol. 1993; 13: 2277-2285Crossref PubMed Scopus (60) Google Scholar) or by a polyguanosine tract, which forms a tertiary structure impervious to either 5′ → 3′ or 3′ → 5′ exonucleases (8Drager R.G. Girard-Bascou J. Choquet Y. Kindle K.L. Stern D.B. Plant J. 1998; 13: 85-96Crossref PubMed Scopus (94) Google Scholar, 14Drager R.G. Zeidler M. Simpson C.L. Stern D.B. RNA. 1996; 2: 652-663PubMed Google Scholar). Both in vitro and in vivo approaches have shown that chloroplast IR sequences do not efficiently terminate transcription, and those tested include Chlamydomonas atpB(13Stern D.B. Kindle K.L. Mol. Cell. Biol. 1993; 13: 2277-2285Crossref PubMed Scopus (60) Google Scholar, 15Rott R. Liveanu V. Drager R.G. Stern D.B. Schuster G. Plant Mol. Biol. 1998; 36: 307-314Crossref PubMed Scopus (40) Google Scholar). This implies that post-transcriptional 3′ end maturation is required. We have previously shown that atpB mRNA undergoes a two-step maturation process (13Stern D.B. Kindle K.L. Mol. Cell. Biol. 1993; 13: 2277-2285Crossref PubMed Scopus (60) Google Scholar). In the first step, endonuclease cleavage occurs at three consecutive positions 8–10 nt downstream of the mature 3′ ends (Fig. 1A). This cleavage is rapid, and in vitro can be detected within seconds of adding an artificial transcript to a Chlamydomonas chloroplast stromal extract. The second step is 3′ → 5′ exonucleolytic trimming, a slower step that requires about 15 min in vitro. Surprisingly, however, the distal product of endonucleolytic cleavage could not be detected in vitro, even when theatpB endonuclease cleavage site (ECS) was placed between two stem-loops or when the artificial transcript was labeled at its 3′ end. This suggested that some mechanism rapidly degraded the downstream cleavage product, unlike the unprocessed pre-mRNA or maturedatpB fragment, both of which were relatively stable in thein vitro system. The atpB mRNA itself is quite stable in vivo, with a half-life estimated at 2–10 h depending on growth conditions (16Salvador M.L. Klein U. Bogorad L. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 1556-1560Crossref PubMed Scopus (60) Google Scholar). Here, we show that cleavage at theatpB ECS potentiates rapid degradation of the downstream fragment in vivo, even if it is a coding region or flanked by protective RNA structures. Our data suggest the activity is an endonuclease that acts as part of a targeted RNA recycling mechanism. The strains used in this study were wild-type P17 (12Stern D.B. Radwanski E.R. Kindle K.L. Plant Cell. 1991; 3: 285-297PubMed Google Scholar), which is derived from CC373 (17Shepherd H.S. Boynton J.E. Gillham N.W. Proc. Natl. Acad. Sci. U. S. A. 1979; 76: 1353-1357Crossref PubMed Scopus (58) Google Scholar), the atpB deletion mutant used as a transformation recipient. Cells were grown in TAP medium (18Harris E.H. Boynton J.E. Gillham N.W. Microbiol. Rev. 1994; 58: 700-754Crossref PubMed Google Scholar) under constant fluorescent light. The uidA constructs shown in Fig. 2 were based on pDG2 (19Sakamoto W. Kindle K.L. Stern D.B. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 497-501Crossref PubMed Scopus (86) Google Scholar). This plasmid contains the petD promoter and 5′ UTR translationally fused to the uidA coding region andrbcL 3′ UTR, and the chimeric gene is inserted into the large inverted repeat of the chloroplast genome downstream of theatpB gene (11Blowers A.D. Ellmore G.S. Klein U. Bogorad L. Plant Cell. 1990; 2: 1059-1070PubMed Google Scholar). We previously engineered a BglII site into the petD 5′ UTR at position +25 relative to the mature RNA 5′ end, which lengthens the transcript by 6 nt and is present in pDG2 (19Sakamoto W. Kindle K.L. Stern D.B. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 497-501Crossref PubMed Scopus (86) Google Scholar). The BglII site was used as an insertion point for the various atpB fragments to make thepetD-uidA and petD-aadA (DG and DA) constructs. The 242-bp ECS fragment, which contains the distal half of the atpB 3′ stem-loop, the ECS, and downstream sequences, was amplified from the WT atpB gene using primers NFIR.3 (loop of IR) and DBS5 (see primers listed in TableI). This PCR product was cloned into the ddT-tailed EcoRV site of pBluescript II SK+ and released with BglII (in primer NFIR.3) and XbaI (in primer DBS5). This BglII-XbaI fragment was ligated into the unique BglII site (+25) of pDG2 after bluntingXbaI and the 3′ half of petD BglII, generating the ECS-DG plasmid.Table IOligonucleotide primers used in this studyPrimersOligonucleotide sequenceatpBstartGCCCCAAGGCATTCATADBS1CAAAATTCTCCACCAGCDBS4CCAGAACAAGCATTCTACTTAGTAGGDBS4anti (P3 in Fig. 2C)CCTACTAAGTAGAATGCTTGTTCTGGDBS5XbaI-CTGTATCTCGTGGACGAClar005 (T7 underlined)TAATACGACTCACTATAGGGGGCTCGTGAAGCGGTTATCGlar006 (T3 underlined)AATTAACCCTCACTAAAGGGGTTATTTGCCAACTACCTTANFIR.3BglII-ATTTGGACACCATTAAGTTGNFIR.5BglII-CAGGTAGCCGAAGGGGP1 (ah-rbc13′-atpB)CCCCTTCCCCTTCGGGACGTCCP2 (ah-atpBecs)GGGAAAGGTGCAACTACGTGGGWS11SmaI-ACTGACATATTTATTTATCCGTTAAWS13BglII-TTTTTTAGCATGTAAACATTAGAAATAC8901CTAAAATAATCTGTCCGG8906GACAAGCTTTACATCCC Open table in a new tab The WT atpB 3′ UTR was amplified with primers DBS4 and DBS5, yielding the same 341-bp fragment previously used for in vitro mRNA maturation studies (13Stern D.B. Kindle K.L. Mol. Cell. Biol. 1993; 13: 2277-2285Crossref PubMed Scopus (60) Google Scholar). The insert was cloned and then released with XhoI andXbaI, blunted, and inserted in the blunted BglII site of the petD 5′ UTR. Both orientations were obtained to make plasmids +IRECS-DG and −IRECS-DG. This 679-bp insert was amplified from the plasmid containing tandem Chlamydomonas atpB and spinachpetD 3′ UTRs (Δ8Δ50; Ref. 13Stern D.B. Kindle K.L. Mol. Cell. Biol. 1993; 13: 2277-2285Crossref PubMed Scopus (60) Google Scholar) with primers DBS4 and 8901. The fragment was inserted into the pDG2 BglII site to generate the IRECSIR-DG plasmid. Finally, the 151-bp IR fragment was made from the source plasmid atpBΔ21, a derivative of the WTatpB gene in which 6 nt of the distal stem of theatpB 3′ stem-loop has been deleted, as well as downstream sequences (12Stern D.B. Radwanski E.R. Kindle K.L. Plant Cell. 1991; 3: 285-297PubMed Google Scholar). At the site of the deletion a BglII site was inserted. The atpBΔ21 Chlamydomonas strain accumulates a discrete atpB transcript of the wild-type size, indicating that the short deletion does not significantly comprise RNA stability (13Stern D.B. Kindle K.L. Mol. Cell. Biol. 1993; 13: 2277-2285Crossref PubMed Scopus (60) Google Scholar). The IR insert was amplified from atpBΔ21 with primers DBS4 (XhoI) and 8906 (HindIII), inserted into pBluescript, excised with XhoI and BglII (the Δ21 3′ deletion end point), and inserted into the petD 5′ UTR BglII site after blunting XhoI and the 5′ half of petD BglII, creating plasmid IR-DG. The pDAAD (20Rott R. Drager R.G. Stern D.B. Schuster G. Mol. Gen. Genet. 1996; 252: 676-683PubMed Google Scholar)-based constructs shown in Fig. 3 were made by releasing thepetD promoter and 5′ UTR fragments from ECS-DG and +IRECS-DG with XhoI (upstream of promoter) and SmaI (3′ to the translation start) and using them to replace the equivalentXhoI-SmaI fragments of pDAAD, generating plasmids ECS-DA and +IRECS-DA. WT and CC373 were referenced above. Strain Δ8 has been described previously (12Stern D.B. Radwanski E.R. Kindle K.L. Plant Cell. 1991; 3: 285-297PubMed Google Scholar) and has a complete atpB transcription unit and a 2-kb deletion beginning 87 bp downstream of the IR. At the site of this deletion aBglII site was inserted, and this was used to create the Δ8pG strain. The plasmid was digested at its unique BglII site, and annealed oligonucleotides carrying a G18 motif, a flanking EcoRI RFLP to determine orientation, and sticky ends compatible with BglII (14Drager R.G. Zeidler M. Simpson C.L. Stern D.B. RNA. 1996; 2: 652-663PubMed Google Scholar) were ligated into it. For IRECSpG, an atpB 3′ UTR-pG-containing fragment was amplified from plasmid atpBΔ8pG using primers DBS4 and 8901 and inserted into a ddT-tailed EcoRV site of pBluescript II SK+. This fragment was released with XhoI (in primer DBS4) and BglII (in primer 8901) and inserted into the BglII site of pDG2, after blunting XhoI and the 3′ petD BglII site, creating plasmid IRECSpGDG. RNA was isolated from 10 ml of cells as previously described (21Drager R.G. Higgs D.C. Kindle K.L. Stern D.B. Plant J. 1999; 19: 521-532Crossref PubMed Scopus (57) Google Scholar). For RNA filter hybridizations, 10 μg of total RNA was fractionated in 1.2% agarose, 6% formaldehyde gels, transferred to Hybond-N (Amersham Biosciences, Inc.), and cross-linked by UV irradiation. Double-stranded DNA probes were labeled by random priming (22Feinberg A.P. Vogelstein B. Anal. Biochem. 1983; 132: 6-13Crossref PubMed Scopus (16653) Google Scholar), and filters were prehybridized and hybridized according to Church and Gilbert (23Church G. Gilbert W. Proc. Natl. Acad. Sci. U. S. A. 1984; 81: 1991-1995Crossref PubMed Scopus (7267) Google Scholar). Double-stranded probes were the following: foruidA, an internal EcoRV fragment; forpetD, the complete 362-bp 5′ UTR was amplified using primers WS13 and WS11; for aadA, the coding region was excised from patpX-AAD (24Goldschmidt-Clermont M. Nucleic Acids Res. 1991; 19: 4083-4090Crossref PubMed Scopus (399) Google Scholar); and for atpB, primers DBS1 and atpBstart were used to amplify the 5′ UTR, because this region is not deleted in CC373. For strand-specific RNA probes, filter hybridizations were performed using the Zeta Probe protocol from Bio-Rad (www.bio-rad.com) with the following modifications. Prehybridization and hybridization were at 65 °C for a minimum of 3 h and 8 h, respectively. All probes were made by linearizing DNA templates and transcribing with T3 or T7 RNA polymerase in the presence of [α-32P]UTP (25Stern D.B. Gruissem W. Cell. 1987; 51: 1145-1157Abstract Full Text PDF PubMed Scopus (298) Google Scholar). The following probes were used. For uidA the template was pBGEV (26Higgs D.C. Colbert J.T. Plant Cell Rep. 1993; 12: 445-452PubMed Google Scholar). Antisense transcripts were made with T7 followingHindIII digestion, and sense transcripts were made with T3 following EcoRI digestion. For aadA the template was amplified from patpXAAD with primers lar005 (beginning with a T7 promoter; sense transcripts) and lar006 (beginning with a T3 promoter; antisense transcripts). For probes spanning the region downstream of the atpB 3′ UTR, primers NFIR.3 and 8901 (minus added restriction sites) were modified to be preceded with T7 and T3 promoters, respectively. T3 transcription produced an antisense transcript, and T7 transcription a sense transcript. All RNA hybridizations were analyzed using the Storm system (Molecular Dynamics Inc., Sunnyvale, CA). For primer extension, 2 ng of primer DBS4anti (P3 in Fig. 2C) were labeled with [γ-32P]ATP and polynucleotide kinase and mixed with 20 μg of Chlamydomonas RNA in a final volume of 8 μl The reaction also contained 0.7 μl of 10 mm dNTPs and 0.7 μl of 10× reaction buffer (10× buffer is 100 mmTris-HCl, pH 8.5, 60 mm MgCl2, 500 mm KCl, 10 mm dithiothreitol). The reaction was incubated for 5 min at 75 °C and 5 min at 50 °C, 4.5 units of avian myeloblastosis virus reverse transcriptase (Promega) were added, and incubation was continued for 15 min at 50 °C. 5 μl of sequencing dye was then added, and samples were analyzed by denaturing polyacrylamide gel electrophoresis. Degradation of downstream sequences is associated with ECS cleavage (Fig. 1A) and may be associated with multiple mechanisms (Fig. 1B). Mutagenesis of the ECS showed, however, that cleavage occurred in vivoeven after multiple mutations were introduced (27Rott R. Liveanu V. Drager R.G. Higgs D.C. Stern D.B. Schuster G. Plant Mol. Biol. 1999; 40: 676-686Crossref Scopus (11) Google Scholar). This suggested that the ECS alone did not determine the processing site. To determine which RNA element(s) might be required for ECS downstream product destabilization, a series of reporter gene constructs were assembled and introduced into Chlamydomonas chloroplasts by biolistic transformation, as shown in Fig.2A. These strains contained a chimeric petD promoter/5′ UTR-uidA(β-glucuronidase) coding region-rbcL 3′ UTR reporter gene, which was located downstream of and in opposite orientation to the endogenous atpB gene. This insertion site has been used for numerous uidA gene fusions in the past (10De Klein U. Camp J.D. Bogorad L. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 3453-3457Crossref PubMed Scopus (61) Google Scholar, 11Blowers A.D. Ellmore G.S. Klein U. Bogorad L. Plant Cell. 1990; 2: 1059-1070PubMed Google Scholar, 19Sakamoto W. Kindle K.L. Stern D.B. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 497-501Crossref PubMed Scopus (86) Google Scholar) and does not affect the expression of atpB, which is used as the selectable marker for transformation of the nonphotosynthetic recipient strain CC373 (see “Experimental Procedures”). The reference strain, DG2, has the structure shown at thetop of Fig. 2A. Five additional strains were created that have insertions at the +25 position relative to the matureuidA mRNA 5′ end. We have previously shown that deletions up to +25 do not affect transcription rates (19Sakamoto W. Kindle K.L. Stern D.B. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 497-501Crossref PubMed Scopus (86) Google Scholar) and that small mutations or a polyguanosine insertion within the +1–25 region do not decrease petD RNA levels (21Drager R.G. Higgs D.C. Kindle K.L. Stern D.B. Plant J. 1999; 19: 521-532Crossref PubMed Scopus (57) Google Scholar, 28Higgs D.C. Shapiro R.S. Kindle K.L. Stern D.B. Mol. Cell. Biol. 1999; 19: 8479-8491Crossref PubMed Scopus (72) Google Scholar). In addition, the sequences around this position are not important for petDmRNA 5′ end maturation (28Higgs D.C. Shapiro R.S. Kindle K.L. Stern D.B. Mol. Cell. Biol. 1999; 19: 8479-8491Crossref PubMed Scopus (72) Google Scholar), although any insertion into this site translationally inactivates the message for unknown reasons. For this reason, analyses were restricted to RNA accumulation. The +25 insertions are shown symbolically in Fig. 2A and include the following sequences: ECS, a 242-bp fragment, which lacks one stem of the atpB 3′ IR, and includes the ECS; +IRECS, a 341-bp fragment, which includes the full 3′ IR and ECS in the sense orientation relative to uidA; −IRECS, which includes the 3′ IR and ECS in the antisense orientation relative to uidA; IRECSIR, a 679-bp fragment, which contains the atpB 3′ IR and ECS, with a second IR derived from the spinach chloroplastpetD gene immediately downstream; and IR, a 151-bp fragment, which has the nearly complete atpB 3′ IR but lacks the ECS. Homoplasmic transformants were obtained, and uidA mRNA was examined with strand-specific probes, with petD mRNA as a loading control. Fig. 2B (top panel) shows results with a probe that detects sense uidA mRNA. As expected, no hybridization occurs with RNA from wild-type cells or CC373, the transformation recipient. On the other hand, a transcript of 2.5 kb accumulates in DG2; this mRNA has a 5′ end at position +1 in the petD 5′ UTR (19Sakamoto W. Kindle K.L. Stern D.B. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 497-501Crossref PubMed Scopus (86) Google Scholar) and terminates in the rbcL3′ UTR at a stem-loop structure (29Dron M. Rahire M. Rochaix J.D. J. Mol. Biol. 1982; 162: 775-793Crossref PubMed Scopus (179) Google Scholar). Among the five newly created strains, three accumulated a sense uidA transcript. Of these, the lowest accumulation relative to petD was in ECS, with higher amounts in −IRECS and IR. The sizes of the chimericuidA transcripts in these strains are slightly larger than in DG2, commensurate with the sizes of insertions at +25. We thus infer that these transcripts have been correctly 5′ end-processed. On the other hand, neither +IRECS nor IRECSIR accumulated senseuidA mRNA. These are also the only two strains that have both the atpB IR and ECS at +25. Because the IR is not responsible for transcription termination, for example it is present in strain IR (see also Refs. 13Stern D.B. Kindle K.L. Mol. Cell. Biol. 1993; 13: 2277-2285Crossref PubMed Scopus (60) Google Scholar and 15Rott R. Liveanu V. Drager R.G. Stern D.B. Schuster G. Plant Mol. Biol. 1998; 36: 307-314Crossref PubMed Scopus (40) Google Scholar), it can be concluded that the IR-ECS combination prevents accumulation of uidA mRNA through RNA degradation. We would explain accumulation in strain ECS by inefficient cleavage in the absence of the IR; uncleaved RNA accumulates, whereas RNAs that are cleaved undergo downstream degradation. The bottom panel of Fig. 2B shows results with a probe designed to detect antisense uidA mRNA. Surprisingly, hybridization was detected in four strains, most prominently in +IRECS and most weakly in ECS (the band marked with anasterisk is an rRNA hybridization artifact). Based on their sizes, these RNAs had 5′ ends in or near the intergenic region betweenatpB and the 3′ end of the uidA cassette and 3′ ends at the IR sequences inserted at the petD +25 position (see Fig. 5 and “Discussion”). We have already shown that theatpB 3′ IR can stabilize RNAs when it is in the antisense orientation (20Rott R. Drager R.G. Stern D.B. Schuster G. Mol. Gen. Genet. 1996; 252: 676-683PubMed Google Scholar). To map the 5′ ends, we selected +IRECS and IRECSIR and used primer extension with RNA from untransformed cells as a negative control. Three primers were tested, as shown at thetop of Fig. 2C (P1, P2, P3). Using P1, no transformant-specific products were detected; however using P2, high molecular weight products were seen for +IRECS and IRECSIR but not for the WT control (data not shown). To map the end precisely, we used the 26-nt primer P3, which anneals upstream of the atpB stop codon. 5′ Ends mapping 6 nt upstream of the primer were found in both transformants, whereas this product was not seen in the control. The additional bands probably result from abortive extension of the abundant endogenous atpB transcript, which extends 1.9 kb upstream and accumulates in all strains. The mapped 5′ ends in +IRECS and IRECSIR lie 72 nt upstream of the atpB stop codon and may represent a normal cleavage site during atpB mRNA degradation. We hypothesize that they accumulate in these transformants due to their stabilization at the 3′ end by IR structures and for unknown reasons, perhaps structural, are not subject to cleavage at the ECS. Note that the IRECSIR antisense transcript is slightly longer (Fig. 2B), in agreement with the fact that it has apetD 5′ UTR insert 338 bp longer than that of +IRECS, which would be included at the 3′ end of the antisense transcript. In wild-type cells no antisense transcript accumulated, based on primer extension. Assuming that the same atpB-internal cleavage occurs in wild-type cells, a short transcript would be generated also flanked by a 3′ IR. However, this hypothetical transcript would only be 168 nt and might be unstable and/or subject to loss during RNA isolation. The lack of an antisense transcript in strain IR, on the other hand, was unexpected given its accumulation in the other IR-containing strains. We note that the +25 insertion in IR is not the entire stem-loop, as it lacks the last 6 of 20 nt in the second stem (12Stern D.B. Radwanski E.R. Kindle K.L. Plant Cell. 1991; 3: 285-297PubMed Google Scholar). Given its AT-richness, the antisense IR sequence may form too weak of a secondary structure to stabilize the uidAantisense transcript. To confirm that the results shown in Fig. 2 were not due to a peculiarity of the reporter gene chosen, certain constructs were replicated using theaadA coding region instead of uidA. aadA is a commonly used selectable marker gene in Chlamydomonaschloroplasts and has also been used to study RNA ciselements (8Drager R.G. Girard-Bascou J. Choquet Y. Kindle K.L. Stern D.B. Plant J. 1998; 13: 85-96Crossref PubMed Scopus (94) Google Scholar, 30Zerges W. Rochaix J.D. Mol. Cell. Biol. 1994; 14: 5268-5277Crossref PubMed Scopus (82) Google Scholar). Fig. 3 shows results from the reference strain, DAAD (petD-aadA), and two derivatives, ECS-DA and +IRECS-DA. These are in every way analogous to the uidA-based constructs DG2, ECS-DG, and +IRECS-DG, respectively. When a dsDNA probe recognizing the aadA coding region was used, a 1.5-kb transcript accumulated in DAAD, as expected. This is the chimeric petD-aadA-rbcLmessage. A 1.7-kb species accumulated in ECS-DA, and a 3.4-kb transcript accumulated in +IRECS-DA. To differentiate between sense and antisense transcripts, a strand-specific probe was used, which only detected antisense aadA mRNAs (bottom panel). This identified the 3.4-kb species but not the 1.5-kb and 1.7-kb transcripts. Thus, a sense aadA transcript accumulates to a low level when the ECS alone is inserted at position +25, not at all when both the IR and ECS are present, and to an increased level when neither is present. On the other hand, the antisense message only accumulated when a stem-loop-forming sequence was present at position +25. These results are entirely consistent with the results foruidA constructs (except the lack of a minor antisense transcript in ECS-DA) and indicate that the reporter gene coding region did not have a significant effect on the outcome. The mechanism by which ECS cleavage potentiates downstream degradation could involve one or a combination of ribonuclease activities known to exist in Chlamydomonaschloroplasts, as shown in Fig. 1B. 5′ → 3′ Exonuclease activity (i) is known from studies of mutants in which particular chloroplast transcripts are unstable due to lack of protection of the 5′ UTR by nucleus-encoded factors and can be blocked by a polyguanosine (pG18) sequence (8Drager R.G. Girard-Bascou J. Choquet Y. Kindle K.L. Stern D.B. Plant J. 1998; 13: 85-96Crossref PubMed Scopus (94) Google Scholar, 31Nickelsen J. Fleischmann M. Boudreau E. Rahire M. Rochaix J.-D. Plant Cell. 1999; 11: 957-970
Referência(s)