Plant Cell, Vol. 11, 1565-1578, August 1999, Copyright © 1999, American Society of Plant Physiologists
Polyadenylation Occurs at Multiple Sites in Maize Mitochondrial cox2 mRNA and Is Independent of Editing Status
D. Shelley Lupolda,b,
Angelina G. F. S. Caoileb, and
David B. Sternb
a Section of Genetics and Development, Cornell University, Ithaca, NY 14853
b Boyce Thompson Institute for Plant Research, Cornell University, Ithaca, NY 14853-1801
Correspondence to:
David B. Stern, ds28{at}cornell.edu (E-mail), 607-255-6695 (fax)
 |
ABSTRACT |
|---|
Polyadenylation of nucleus-encoded transcripts has a well-defined role in gene expression. The extent and function of polyadenylation in organelles and prokaryotic systems, however, are less well documented. Recent reports of polyadenylation-mediated RNA destabilization in Escherichia coli and in vascular plant chloroplasts prompted us to look for polyadenylation in plant mitochondria. Here, we report the use of reverse transcription–polymerase chain reaction to map multiple polyadenylate addition sites in maize mitochondrial cox2 transcripts. The lack of sequence conservation surrounding these sites suggests that polyadenylation may occur at many 3' termini created by endoribonucleolytic and/or exoribonucleolytic activities, including those activities involved in 3' end maturation. Endogenous transcripts could be efficiently polyadenylated in vitro by using maize mitochondrial lysates with an activity that added AMP more efficiently than GMP. Polyadenylated substrates were tested for stability in maize mitochondrial S100 extracts, and we found that, compared with nonpolyadenylated RNAs, the polyadenylated substrates were less stable. Taken together with the low abundance of polyadenylated RNAs in maize mitochondria, our results are consistent with a degradation-related process. The fact that polyadenylation does not dramatically destabilize plant mitochondrial transcripts, at least in vitro, is in agreement with results obtained for animal mitochondria but differs from those obtained for chloroplasts and E. coli. Because fully edited, partially edited, and unedited transcripts were found among the cloned polyadenylated cox2 cDNAs, we conclude that RNA editing and polyadenylation are independent processes in maize mitochondria.
|
|
INTRODUCTION |
|---|
Present-day mitochondria have almost certainly evolved from a prokaryotic endosymbiont (reviewed in Gray 1992 
). These organelles possess their own genomes and gene expression machinery; however, during evolution, most of the genetic information of the mitochondrial ancestor was transferred to the nuclear genome. Mitochondrial genes in Saccharomyces cerevisiae and metazoans are transcribed by a nucleus-encoded T7-like RNA polymerase and accessory factors (reviewed in Tracy and Stern 1995 ), and candidate plant nuclear genes encoding mitochondrial RNA polymerase have been identified (Cermakian et al. 1996 ; Hedtke et al. 1997 ; Young et al. 1998 ; Chang et al. 1999 ). In plants, promoter strength may play a regulatory role in gene expression (Mulligan et al. 1991 ), but post-transcriptional regulation also can occur by differential RNA stability (Finnegan and Brown 1990 ).
The maize mitochondrial genome is typical of those found in vascular plants. It can be genetically mapped as a single circular molecule of 570 kb, with multiple repeated sequences giving rise to a variety of stably inherited subgenomic recombination products (Lonsdale et al. 1984 ). As in other species, maize mitochondrial primary transcripts are subject to both cis-splicing (e.g., Fox and Leaver 1981 ) and trans-splicing (Pereira de Souza et al. 1991 ); in many cases, RNA editing, usually by conversion of cytosine to uridine, occurs widely and is often required to form an initiation or stop codon or to create a functional protein (reviewed in Maier et al. 1996 ).
3' end formation and regulation of mRNA stability are poorly understood in plant mitochondria. Recent studies have shown that as in chloroplasts, 3' untranslated region (UTR) sequences do not serve as transcription terminators but instead direct 3' end maturation (Dombrowski et al. 1997 ). The role of 3' UTR sequences and structures in RNA stability has been demonstrated in the cases of rice cob, in which two genes differing only in their 3' UTRs have widely different stabilities (Kaleikau et al. 1992 ), and rapeseed cytoplasmic male sterility locus open reading frame 138, in which similar effects have been documented both in vivo and in vitro (Bellaoui et al. 1997 ).
In this article, we report evidence for extensive polyadenylation of maize mitochondrial cox2 transcripts. Although polyadenylation of nuclear and some bacterial and metazoan mitochondrial transcripts is well known (Gelfand and Attardi 1981 ; Ojala et al. 1981 ; Wahle and Keller 1992 ; Sarkar 1997 ), data demonstrating polyadenylation of plant mitochondrial mRNAs had not been reported to our knowledge. The recent rediscovery of polyadenylation in higher plant chloroplasts and its role in RNA degradation (Kudla et al. 1996 ; Lisitsky et al. 1996 ; reviewed in Hayes et al. 1999 ) prompted us to look for polyadenylation in plant mitochondria. We report that polyadenylation occurs at many sites in the 3' region of maize cox2 mRNA, apparently at 3' termini created by endonucleolytic and/or exonucleolytic cleavage events, including 3' end maturation. Although polyadenylation appears not to dramatically destabilize plant mitochondrial transcripts in vitro, the low abundance of these transcripts is consistent with their association with an RNA degradation pathway. Our results emphasize that polyadenylation is a widespread process in the DNA-containing organelles of eukaryotic cells; however, its metabolic role may vary.
|
|
RESULTS |
|---|
Detection of Polyadenylated mRNA in Maize Mitochondria and Identification of Poly(A) Addition Sites
We selected cox2 for our initial studies of polyadenylation. cox2 encodes subunit II of the cytochrome oxidase complex and was the first plant mitochondrial gene sequenced (Fox and Leaver 1981 ). To look for polyadenylated cox2 transcripts, we reverse transcribed RNA isolated from purified maize mitochondria (mtRNA) by using an oligo(dT)17 adapter oligonucleotide, and the resulting cDNAs were polymerase chain reaction (PCR) amplified with an adapter primer and a second primer located within the cox2 coding region (primer ZMC22; Figure 1A). Two major size classes of PCR products were purified from an agarose gel and cloned into a plasmid vector. These products were not observed using the oligo(dT) primer alone, in the absence of reverse transcriptase, or when RNA preparations were treated with oligo(dT) and RNase H, which should specifically degrade the poly(A) tails of mRNAs (Figure 1B). Positive cDNAs were detected by colony hybridization using the SnaBI-AvaI fragment of the cox2 3' region (see Figure 1A), and 60 positive transformants were selected for further analysis. Most of the positive clones fell into two classes containing ~400- or 700-bp inserts.
View larger version (55K):
[in this window]
[in a new window]
|
Figure 1.
Isolation of cDNAs Corresponding to Polyadenylated cox2 mRNAs.
(A) The cox2 exon 2 and downstream region of the maize mitochondrial genome. Shown is a map of the clones, relevant restriction sites, primers (indicated by dashes), and probes (the direction of the 3' end mapping [3' EM] probe is indicated by an arrow) used in the experiments (see Methods). The arrow labeled N6 represents the DNA insert in the N6 clone. The other arrows connect labels with the corresponding objects.
(B) Filter hybridization analysis of RT-PCR products. Simultaneous reactions were performed with the oligo(dT) (dT) primer alone or with the oligo(dT) primer as well as the specific cox2 primer, one of which lacked reverse transcriptase (no RT). In some cases, the mtRNA was pretreated with various combinations of oligo(dT) and RNase H; the dash column indicates reactions in which neither was added. The products were separated in an agarose gel, which was blotted and hybridized with the colony screening probe shown in (A). The arrows indicate the two major size classes of PCR products obtained, and the markers were  X174 DNA digested with HinfI.
(C) Dideoxy sequencing reactions of two clones having the two most-represented polyadenylation sites, with A36 and A33 giving the deduced lengths of the poly(A) tails. Clones 7-1R and 20-1R were sequenced using the M13 reverse primer.
|
|
Both strands of each positive clone were sequenced, and as expected, each contained a 3' polyadenosine tract of variable length. A representative sequencing gel of two clones having the two most represented poly(A) sites is shown in Figure 1C. Like the vast majority of positive clones, the poly(A) tract was homopolymeric (see below). Figure 2A shows the locations of the 23 poly(A) sites identified within the cox2 coding region and 3' UTR. Of the 60 clones sequenced, half were polyadenylated at apparently random sites between 138 nucleotides upstream and 377 nucleotides downstream of the stop codon. The two most heavily represented sites are indicated by arrows and are located 41 (site 7) and 363 (site 20) nucleotides downstream of the stop codon, respectively, and they accounted for the remaining clones. The 322 bp between these sites correspond well to the difference between the sizes of the two major reverse transcription (RT)–PCR products shown in Figure 1B. Site 7 was found 13 times, and site 20 was found 16 times. All other sites were found once, with the exception of the sites indicated by a double-length boldface line, which were found two (sites 8 and 22) or three (sites 11, 19, and 21) times.
View larger version (39K):
[in this window]
[in a new window]
|
Figure 2.
Polyadenylation of cox2 Transcripts.
(A) Map showing the location and relative frequency of the 23 polyadenylation sites for cox2 found in oligo(dT)-primed cDNAs. Exon 2 of cox2 is indicated by a shaded box. The mature 3' end is indicated and coincides with the region including sites 18 to 22. The sites fell into three categories: those with one representative clone, shown by a numbered vertical line; those with two or three representative clones, shown by a numbered boldface vertical line; and those with 13 to 16 representative clones, shown by a numbered boldface line ending in an arrow.
(B) A histogram showing the tail nucleotide sequences found and their frequencies. Question marks indicate the presence of ambiguous sequences. One clone lacked the adapter and polyadenosine sequences altogether.
|
|
We examined the sequences found upstream and downstream of each polyadenylation site and the number of clones having each site. No consensus sequence was identified within the 40 nucleotides surrounding the polyadenylation sites; however, in 20 clones (33%), adenosines were found at the junction between the upstream and downstream sequences, making it impossible to determine precisely where polyadenylation occurred. There was also no obvious similarity either in sequence or in base composition between the two most heavily used sites, 7 and 20 (see also Figure 3B). This suggested that the frequency of polyadenylated RNAs might depend on the relative levels of cox2 mRNAs having the respective 3' termini (see below).
View larger version (26K):
[in this window]
[in a new window]
|
Figure 3.
Major 3' Ends and Polyadenylation Sites in the cox2 3' UTR.
(A) Determination of the mature 3' end of cox2 transcripts by S1 nuclease protection. The probe used was the 3' end–labeled SphI-NdeI restriction fragment of N6SB (see Figure 1A). The amount of probe used (1x or 2.5x) and the addition (+) or not (-) of mtRNA, yeast tRNA, and/or S1 nuclease are indicated. Size markers were HinfI-digested X174 and an unrelated sequence ladder (data not shown). The protected fragments map from 360 to 365 nucleotides (nt) downstream of the stop codon. The self-annealed double-stranded probe is 270 bp.
(B) Sequences surrounding sites 7 and 20 in the cox2 3' UTR. The heterogeneous 3' ends are marked by arrows above the sequence, and poly(A) addition sites are marked by asterisks below the sequences, with the number indicating the number of cDNAs recovered for each site.
|
|
Figure 2B shows the nucleotide composition of the tails sequenced and how many clones had each tail type. Because the oligo(dT) primer can anneal anywhere within a poly(A) tail, these results are likely to underestimate the length of poly(A) tails, which for a few clones were apparently less than the length of the oligo(dT) primer (17 nucleotides). The randomly selected clones may also not reflect the composition of tails in vivo. Of the clones sequenced, 49 (80%) were homopolymeric for adenosines and ranged in size from 14 to 36 nucleotides, with the majority having 16 to 20 adenosines. Three clones had a single guanosine embedded in their tails, one had two guanosines separated by an adenosine, three had a single uridine at or near their 3' ends, and several had interspersed uridines or cytosines. Overall, the average tail length was 22 nucleotides, with a range of 14 to 52 nucleotides. Taken together, these results appear to reflect an ability of the polyadenylation machinery to utilize inefficiently any of the other three nucleotides.
A Major Poly(A) Addition Site Corresponds to the Mature 3' End of the cox2 Transcript
The preponderance of clones with poly(A) added at either site 7 or site 20 raised the possibility that one or both of these sites might correspond to abundant termini of cox2 mRNA. To map the mature 3' ends of the cox2 transcript, we performed S1 nuclease protection assays. The SphI-NdeI fragment shown in Figure 1A was labeled on the antisense strand and hybridized with mtRNA, or with tRNA as a control. The resulting DNA–RNA hybrids were subjected to S1 nuclease digestion and sized relative to a DNA sequence ladder and other size markers. Figure 3A shows that the major products were 139 to 144 nucleotides long, corresponding to positions 360 to 365 nucleotides downstream of the stop codon, with the predominant length being 142 nucleotides. Poly(A) addition sites 18 to 22 lie within this six-nucleotide region, accounting for polyadenylation sites in 25 (42%) of the 60 clones (Figure 3B). Thus, it appears that mature cox2 mRNA is a common substrate for poly(A) addition. In contrast, S1 protection revealed only a minor protected product corresponding to a 3' end at or near site 7 (D.S. Lupold and D.B. Stern, unpublished data; see Figure 3B for the sequence surrounding site 7). This suggests that polyadenylation may occur at site 7 as part of an RNA degradation pathway or that RNA processed at site 7 might be stabilized by polyadenylation.
A Maize Mitochondrial Protein Extract Possesses Polyadenylation Activity
To detect presumed poly(A) addition activity in maize mitochondria, we isolated a total soluble fraction from purified, intact mitochondria that were lysed in the presence of 0.5% Triton X-100 and 1 M KCl. Because endogenous RNA had not been removed from these lysates, we performed an experiment to see whether poly(A) addition activity could utilize these natural substrates. To do this, we incubated the lysates in the presence of 0.5 mM unlabeled ATP as well as a small amount of 
-32P-ATP. As shown in lanes 2 to 6 of Figure 4, a time-dependent incorporation of label was observed, yielding an increasing amount of a high molecular weight heterodisperse product as well as a discrete product of ~75 nucleotides. The discrete product almost certainly corresponds to tRNA, with ATP added by a tRNA nucleotidyltransferase activity; such activities are known to be present in plant mitochondrial extracts (Hanic-Joyce and Gray 1990 ). Another discrete product of ~120 nucleotides may correspond to the 5S ribosomal RNA (Stern et al. 1982 ). To verify that the heterodisperse products consisted of RNA, we treated aliquots with RNase A or DNase I and found them to be RNase sensitive but DNase I resistant (lanes 7 and 8, respectively).
View larger version (101K):
[in this window]
[in a new window]
|
Figure 4.
In Vitro Polyadenylation in a Maize S100 Extract.
Preparation of extracts and reaction conditions are described in Methods. The various additions or treatments used are indicated at the top of the gel. (+) indicates that the reagent was used, whereas (-) indicates that it was not used. Numbers at left indicate size markers (lane 1; marked M at the top) in nucleotides, whereas numbers across the top represent reaction times in minutes. Electrophoresis was in a 6% denaturing polyacrylamide gel. u, units.
|
|
Because cDNA sequences had shown low amounts of non-adenosine nucleotides in the post-transcriptionally added tails, the mitochondrial lysates were tested for their abilities to add GMP. Lanes 9 to 12 of Figure 4 show that when 0.5 mM GTP and a small amount of 32P-GTP were used, there was little incorporation. Quantification with a PhosphorImager suggested that the amount incorporated was ~0.1% of the ATP/ATP level. However, when 32P-GTP was provided in the presence of 0.5 mM ATP, incorporation was observed at ~1% of the ATP/ATP level (lanes 12 to 14). The labeled material was sensitive to RNase T1 (Figure 4, lane 15), indicating that GMP was in fact incorporated into RNA rather than being converted into another nucleotide. Taken together, the results shown in Figure 4 strongly suggest that plant mitochondria contain a poly(A) polymerase activity that is responsible for generating the cox2 transcripts that were amplified and sequenced. Whereas poly(A) polymerase activity is also found in plant nuclei and chloroplasts, the purification procedure used for maize mitochondria should result in undetectable levels of contamination (Moore and Proudlove 1983 ). In addition, we did not observe the rapid degradation of polyadenylated transcripts in these same protein extracts that would have been expected from plastid contamination (see below). Finally, the results shown in Figure 4 suggest that although ATP is the preferred substrate for this terminal transferase activity, GTP can be incorporated at a low rate. This finding is fully consistent with the cDNA sequences shown in Figure 2.
Effect of Polyadenylation on Transcript Degradation in Maize Mitochondrial Extracts
Because polyadenylation could have a positive, neutral, or negative effect on mRNA stability, experiments were performed to compare the degradation rates of in vitro–synthesized transcripts possessing or lacking poly(A) tails. Two different substrates were used for these experiments, as shown in Figure 5A, consisting of the mRNA-like strands and ending at site 7 or site 20, which were the most frequently encountered sites of poly(A) addition. Site 7 is relatively rare as a 3' end in the steady state cox2 mRNA population, whereas site 20 is abundant (Figure 3). Apart from three guanosines from the T7 promoter included in one of the PCR primers used to amplify the fragments used for RNA synthesis, the substrates contained only cox2 sequences. To test the effect of poly(A) on transcript stability, we made a second version of each substrate in which 25 adenosines were added at the 3' end. This is similar to the average number of adenosines found in the cDNAs we sequenced (Figure 2B).
View larger version (55K):
[in this window]
[in a new window]
|
Figure 5.
In Vitro RNA Degradation in a Maize Mitochondrial S100 Extract.
(A) Map of the cox2 3' UTR, showing to scale the locations, polarity, and sizes of substrates. The 203-nucleotide site 7 substrate (228 nucleotides with the poly[A] tail) did not contain any known editing sites, whereas the 193-nucleotide site 20 substrate (218 nucleotides with the poly[A] tail) contains one editing site at position 1603 of the cox2 sequence; this site was unedited in the substrate used, as indicated by the open circle.
(B) and (C) Six percent denaturing polyacrylamide gel analysis of reaction products at the times indicated. The numbers below the gels indicate the percentage of each RNA that remained, with the zero time point set to 100%. The values shown are for these gels; repetitions of each experiment were in good agreement. For example, for site 7 RNA, an average of 81% of the nonpolyadenylated substrate remained after 90 min when incubated alone or 67% when incubated in the presence of the poly(A)+ substrate. An average of 30% of the poly(A)+ substrate remained after 90 min when incubated alone or 49% when incubated in the presence of the poly(A)- substrate. Numbers across the top indicate reaction times in minutes; +p(A)25 indicates that the substrate included a 25-nucleotide poly(A) tail; -p(A) indicates that the substrate lacked a poly(A) tail. The reaction contents are indicated at the top of each gel; each substrate was added alone, or the two were added simultaneously (mixed) in approximately equal amounts. Below each gel, the percentage remaining is indicated; the top row is for the +p(A)25 transcript, where present, and the lower row is for the -p(A) transcript, where present.
|
|
Equal counts of the 32P-labeled transcripts with or without poly(A) tails were incubated alone or mixed together in maize mitochondrial S100 extracts for varying lengths of time, and the products were analyzed by gel electrophoresis. Figure 5B shows results for site 7 RNA. The nonpolyadenylated transcript decayed slowly, with 84% remaining at the end of the 90-min incubation period. The polyadenylated site 7 RNA, however, decayed faster, with only 20% remaining after 90 min. When mixed together, 61% of the nonpolyadenylated transcript was detectable after 90 min, whereas 44% of the polyadenylated transcript remained. This pattern was reproducible in several experiments and by using independent S100 preparations, with greater differences always seen when the transcripts were incubated alone rather than mixed together. This effect might be due to competition for an RNase activity and, interestingly, contrasts with what was seen during in vitro studies of chloroplast RNA polyadenylation-mediated decay, where the enhanced instability of poly(A)-containing transcripts was most evident when mixed with their nonpolyadenylated counterpart (Lisitsky et al. 1997 ).
Results using site 20 RNA are shown in Figure 5C, and they resemble those obtained for site 7. Because site 20 coincides with the major 3' end of cox2 mRNA, we anticipated that a processing mechanism might remove the poly(A) tail, regenerating a poly(A)- molecule. Indeed, a small amount of a product of this size accumulated when the poly(A)-containing substrate was incubated alone. However, the majority of the poly(A)-containing substrate did not appear to accumulate as this product, suggesting that processing is inefficient, the product is unstable, and/or the processing machinery competes poorly with the RNA degradation machinery. Instead, many faster migrating bands are seen that could represent pauses in a putative exonucleolytic degradation mechanism. Taken together, our results suggest that in this in vitro system, transcripts terminating in 25 adenylate residues are less stable than their nonpolyadenylated counterparts.
RNA Editing and Polyadenylation Appear to Be Largely Independent Processes
Plant mitochondrial RNAs are subject to extensive post-transcriptional C-to-U editing, often resulting in codon changes required to specify functional proteins (reviewed in Maier et al. 1996 ). It was possible that polyadenylation was a mechanism to promote degradation of incorrectly or partially edited transcripts in maize mitochondria. The maize mitochondrial cox2 mRNA is known to be edited (Yang and Mulligan 1991 ), and six of these editing sites were present in cox2 sequences analyzed for poly(A) addition. Therefore, the editing patterns of the polyadenylated clones were determined, and the results are shown in Figure 6A. Of the 60 cDNAs, 32 (53%) were fully edited, 13 (22%) were unedited, and 15 (25%) were partially edited. Each of the six sites was edited ~70% of the time, with the exception of site 2, which was edited 58% of the time.
View larger version (22K):
[in this window]
[in a new window]
|
Figure 6.
Editing Status of the Polyadenylated cDNAs.
(A) Filled circles represent sites that were edited in the sequenced clones, and open circles represent unedited sites. The number of polyadenylated clones having each pattern is indicated at left. Numbers at top represent editing sites.
(B) Editing status of bulk mtRNA. The sequencing gels show heterogeneity at each editing site; these sites are indicated by numbered asterisks, and the corresponding sequences and codons are indicated below the gels.
(C) Editing frequency of bulk RNA was determined by using a PhosphorImager to quantify the edited (lanes T) and unedited (lanes C) bands in (B) for each edited position, using nearby unedited T and C bands as controls for variation in band intensity related to sequence length and composition. The values for bulk RNA are averages of three separate experiments. pA, poly (A).
|
|
To test whether polyadenylated mRNAs exhibited a biased editing pattern when compared with the average status of cox2 transcripts, we assessed the editing status of bulk mitochondrial cox2 mRNA. RNA was isolated from intact mitochondria, reverse transcribed using the cox2 3' UTR–specific primer ZMC24 (see Figure 1), and then PCR amplified using ZMC24 and the upstream cox2 coding region–specific primer ZMC25. The product was gel purified and directly sequenced with the internal primer ZMC26. The experiment was performed in triplicate; a representative gel is shown in Figure 6B. The bands representing the edited (lanes T) and unedited (lanes C) bases for each edited position were quantified using a PhosphorImager.
Figure 6C shows that the degree of editing in the cDNA clones closely paralleled the degree of editing in bulk mitochondrial mRNA in most cases. However, site 2 was consistently edited to a higher degree in bulk mRNA than in the cDNA clones. Sites 1 and 2 are in a single codon (see Figure 6B), and when the second site is unedited, the codon specifies Pro (CCA) or Ser (UCA) instead of the correct Leu. In two-thirds of the partially edited cDNA clones, the first site was edited whereas the second was not, specifying Ser. With bulk mRNA, this editing pattern was largely reversed, leading to a high percentage of Leu (CUA or UUA) codons. Thus, at least in this instance, polyadenylation may have a slight preference for the nonfunctional transcript.
|
|
DISCUSSION |
|---|
Function of Polyadenylation in Maize Mitochondria
Here, we have demonstrated polyadenylation of plant mitochondrial transcripts by mapping numerous poly(A) addition sites in maize cox2 cDNAs. Approximately 40% of the cDNA clones possessed tails at or near the major mature 3' end (sites 18 to 22), and ~25% had tails at an upstream site that is a minor 3' end. Despite several attempts, we have been unable to obtain reliable data regarding the proportion of poly(A)-containing cox2 transcripts among the steady state population; however, given the large number of PCR cycles required to amplify the cDNAs, it is unlikely to be any higher than the several percent estimated for prokaryotic systems (reviewed in Sarkar 1997 ).
Regarding the function of polyadenylation in maize mitochondria, two somewhat conflicting observations were made. The first was that two different polyadenylated synthetic transcripts were only slightly less stable when compared with their nonpolyadenylated counterparts in an in vitro system. When these same transcripts were tested in a chloroplast extract, the poly(A)+ but not poly(A)- transcripts were highly unstable (data not shown), in agreement with other results published for chloroplasts (Kudla et al. 1996 ; Lisitsky et al. 1996 ). Thus, there is nothing inherently stable about the particular sequences that were chosen. The second observation is that poly(A)+ transcripts are of low abundance, as determined by RNase protection, and require sensitive RT-PCR techniques to be amplified as cDNAs. This is consistent with poly(A)-mediated instability or with very inefficient poly(A) addition.
These findings suggest two main alternative interpretations. The first is that poly(A) tails do not confer relative instability to maize mitochondrial RNAs and do not have any other role in gene expression. The second possibility is that polyadenylation is indeed part of a normal RNA degradation or gene regulation pathway and that the relatively minor differential stability in the mitochondrial S100 extract reflects the particular in vitro conditions or substrates used. We believe that the available data support the second possibility for the following reasons.
First, polyadenylation is widespread in mitochondria and thus unlikely to be an evolutionary remnant in plants. For example, apart from the ample documentation in animals cited earlier, oligoadenylation of yeast (Hendler et al. 1975 ; Yuckenberg and Phillips 1982 ; but see Groot et al. 1974 ) and trypanosome (Feagin et al. 1985 ; Bhat et al. 1992 ) mitochondrial transcripts has been reported. Second, whereas polyadenylation of plant mitochondrial transcripts might not confer the striking instability that it does in chloroplasts (Lisitsky et al. 1997 ) and Escherichia coli (Haugel-Nielsen et al. 1996 ), in other mitochondrial systems, poly(A) has still been implicated in RNA decay, for example, in rat, in which there is a correlation between poly(A) tail shortening and mRNA decay (Avadhani 1979 ).
Third, whereas the half-lives of plant mitochondrial transcripts have not been directly measured, in maize, some mitochondrial transcripts that can be labeled by run-on transcription do not appear in the steady state population, implying that there is an active and selective degradation mechanism (Finnegan and Brown 1990 ). Fourth, many of the poly(A) tails we found were not added at the mature 3' end, implying that the tails were added during an RNA degradation process or after abortive transcription termination (Hajnsdorf et al. 1996 ).
Although we did not observe substantial differences in transcript stability using different temperatures or protein preparations (data not shown), it is possible that the S100 extract was depleted of certain ribonuclease activities. Whereas poly(A)-enhanced RNA decay occurs in chloroplast soluble protein extracts, the equivalent factors could be membrane associated in mitochondria. Although we did not test membrane protein fractions for RNase activity, Gagliardi and Leaver 1999 reported little RNase activity in matrix fractions from mitochondria of etiolated sunflower cotyledons, whereas two distinct RNase activities were found in the membrane fractions. In contrast, in vitro studies of mRNA 3' end maturation (Dombrowski et al. 1997 ) and tRNA processing (Marchfelder and Brennicke 1994 ) have utilized mitochondrial lysates much like ours, with lysis in the presence of nonionic detergent followed by clarification of the lysate. We also cannot rule out that tissues other than the etiolated hypocotyls used in our study might express different RNases. For example, the major cytosolic ribonucleases of Arabidopsis accumulate in a tissue-specific and developmentally regulated fashion (Yen and Green 1991 ; Bariola et al. 1994 ).
Determination of Maize Mitochondrial Poly(A) Addition Sites
Because there was no sequence conservation surrounding the poly(A) addition sites, we concluded that polyadenylation occurs in a non-sequence-specific manner, perhaps stochastically at any available 3' end, analogous to the tailing of most or all animal mitochondrial transcripts created by endonucleolytic processing (Hirsch and Penman 1973 ). Similarly, transcripts in prokaryotes, bacteriophage T7, and chloroplasts also exhibit polyadenylation of endonuclease cleavage products without the requirement for specific sequence elements (reviewed in Sarkar 1997 ), although RNA secondary structures may modulate accessibility of the poly(A) polymerase (Xu et al. 1993 ). This contrasts with the case in nuclei of metazoans, yeast, and plants, in which cleavage and polyadenylation are tightly coupled and the site is determined by specific sequence elements (reviewed in Wahle and Keller 1992 ; Rothnie 1996 ; Li and Hunt 1997 ).
To determine whether other maize mitochondrial mRNAs exhibit such heterogeneity in their poly(A) addition sites, we have initiated a follow-up study focusing on cox3. Our preliminary data indicate that 12 of 12 maize cox3 poly(A)-containing cDNAs have the same poly(A) site 56 nucleotides downstream of the stop codon, after UU (D. Gingerich, D.S. Lupold, and D.B. Stern, unpublished data). The identification of additional cox3 poly(A)-containing cDNAs and those for other maize mitochondrial mRNAs is required before we can draw strong conclusions about poly(A) site selection.
Poly(A) Tail Length and Composition
The poly(A) tails of the sequenced cDNA clones ranged from 14 to 52 nucleotides; however, because the oligo(dT) primer can anneal anywhere within a poly(A) tail, these results likely underestimate the length of poly(A) tails in vivo. When unlabeled substrates consisting of endogenous mitochondrial RNAs were in vitro polyadenylated using -32P-ATP, the products were long; however, the substrates were certainly of variable length, and thus the amount of poly(A) addition is difficult to estimate. Based on cDNA sequence data, our results are not conclusive but suggest that in maize mitochondria, poly(A) tails are relatively short, perhaps similar in size to the 57 nucleotides estimated for human mitochondria (Hirsch and Penman 1973 ) and the 35 to 55 nucleotides estimated for rat mitochondria (Avadhani 1979 ) but longer than those in yeast mitochondria (Yuckenberg and Phillips 1982 ) and at the termini of stable E. coli RNAs (Li et al. 1998 ).
The base composition of the tails was examined during the sequencing of the cDNA clones. Eighty percent of the clones were homopolymeric for adenosine, whereas 20% had one or more other nucleotides interspersed, indicating that the poly(A) machinery can inefficiently use any of the other three nucleotides. This finding was supported by the ability of mitochondrial lysates to incorporate limited amounts of guanosine into RNA tails (Figure 4). In this regard, plant mitochondrial polyadenylation most closely resembles that of metazoans, yeast, plants, and prokaryotes and can be distinguished from that of chloroplasts and bacteriophage T7 (Wahle and Keller 1992 ; Lisitsky et al. 1996 ; Sarkar 1997 ).
The possibility of significant contamination of our mitochondrial extracts by nuclear proteins can be largely discounted for several reasons. First, our purification procedure includes both differential centrifugation and density gradient sedimentation, previously shown to yield essentially pure and intact mitochondria (Moore and Proudlove 1983 ). Second, the presence of PCR-amplifiable poly(A)+ cDNAs is consistent with such an activity being mitochondrially localized. Finally, such an activity must obviously be present in other mitochondrial systems where polyadenylation occurs, and indeed, poly(A) polymerase has been partially purified from rat mitochondria (Rose et al. 1975 ).
RNA Editing and Polyadenylation
Because fully edited, partially edited, and unedited transcripts were found among cox2 cDNAs, we concluded that RNA editing and polyadenylation are independent processes, in accord with conclusions reached for Trypanosoma brucei mitochondria (Koslowsky and Yahampath 1997 ). This is not surprising, because there is no defined order to post-transcriptional processes in plant mitochondria; for example, there is no 5' to 3' polarity of editing, and each site has a characteristic editing frequency (Wilson and Hanson 1996 ). Transcript abundance and editing are linked in some cases (e.g., atp6 in petunia; Lu and Hanson 1992 ) but not in others (e.g., atp6 in sorghum; Kempken and Howad 1996 ). Examination of polyadenylation of plant mitochondrial RNAs during development, and for multiple genes in multiple species, is required to more fully elucidate its function. In this regard, newly published data suggest a role for polyadenylation in regulating the stability of a sunflower mitochondrial transcript involved in the expression of cytoplasmic male sterility (Gagliardi and Leaver 1999 ).
|
|
METHODS |
|---|
Seedling Growth, Mitochondrial Isolation, and RNA Extraction
Pioneer brand 3377 maize (Pioneer Hi-Bred International, Johnston, IA) was germinated and grown in the dark for 4 to 5 days. Intact mitochondria were isolated from etiolated maize seedlings as described (Rapp and Stern 1992 ). RNA was extracted from mitochondria as described (Stern and Newton 1986 ).
DNA Clones and Sequences
N6 has the 5.3-kb XhoI fragment from the N5G8 cosmid (kindly provided by Christiane Fauron, University of Utah, Salt Lake City; Lonsdale et al. 1984 ; Fauron and Havlik 1988 ) inserted into pBluescript SK- (Stratagene, La Jolla, CA). N6SB has the SalI-BamHI fragment from N6 subcloned into pBluescript KS+. N6 was sequenced with oligonucleotides ZMC23 and ZMC27 extending from positions 1748 to 1762 (GenBank accession number
V00712) and 541 to 559 bp downstream of the stop codon of the cox2 gene, respectively, to provide additional cox2 3' untranslated region (UTR) sequence information.
PCR Amplification and Identification of Polyadenylation Sites
Mitochondrial RNA (mtRNA; 10 µg) was used as a template for synthesis of oligo(dT)-primed cDNA with the dT-adapter primer, as described previously (Lisitsky et al. 1996 ). This cDNA was polymerase chain reaction (PCR) amplified, with the adapter primer and oligonucleotide ZMC22 extending from position 1338 to position 1354 of the cox2 gene (see Figure 1A). Amplification was performed under standard conditions for 50 cycles of 1 min each at 94, 50, and 72°C, with the addition of extra enzyme (Promega) after 25 cycles. Controls were treated with 0.8 units of RNase H (Gibco BRL), 0.3 µg of oligo(dT)20, or both, as previously described (Brewer and Ross 1988 ). PCR products were size selected (two size classes, as shown in Figure 1B) in an agarose gel and ligated into pUC57/T (MBI Fermentas, Amherst, NY). Colony hybridization with the SnaBI-AvaI fragment of the cox2 3' region was used to select clones for further evaluation. These were sequenced in both directions using vector-specific M13 (-20) and reverse primers to identify polyadenylation sites and check the editing status of each clone. Manganese was added to the sequencing reactions to facilitate the reading of sequences close to the primers.
Determination of the Mature 3' End of the cox2 Transcript
The 270-bp SphI-NdeI restriction fragment was isolated from the N6SB clone and end labeled using T4 DNA polymerase (Promega). Briefly, the T4 DNA polymerase reaction was allowed to proceed for 5 min at 16°C in the absence of deoxynucleotide triphosphates and then labeled with -32P-dCTP in the absence of dGTP for 15 min at 16°C to preferentially label the 3' end of the antisense strand at the SphI site, chased with cold dCTP and dGTP for 5 min at 16°C, and passed over a Sephadex G-25 (Sigma) spin column to remove unincorporated deoxynucleotide triphosphates. S1 nuclease protection was performed as described (Sambrook et al. 1989 ) with two different amounts of probe, either 1x or 2.5x, protected by mtRNA or tRNA.
Protein Extract Preparation, in Vitro Polyadenylation, and Degradation Assays
Intact mitochondria were lysed in the presence of 1 M KCl and 0.5% Triton X-100 (Sigma) followed by centrifugation of the lysate at 100,000g. Endogenous RNA in the S100 extract (450 µg of protein per reaction) was polyadenylated at 30°C in the presence of poly(A) polymerase buffer (40 mM Tris-HCl, pH 7.9, 2.5 mM MnCl2, 250 mM NaCl, 1 mM EDTA, and 10 mM MgCl2) with the addition of labeled or unlabeled nucleotide triphosphates, as indicated in Figure 4. Reactions were stopped by incubation at 37°C for 15 min in the presence of 0.3 M NaCl, 10 mM Tris-HCl, pH 7.9, 5 mM EDTA, 0.8% SDS, and 800 µg of proteinase K. Fifty micrograms of yeast tRNA and 500 µM aurintricarboxylic acid were added before organic extractions and ethanol precipitation. Where indicated, treatment with RNase A, RNase T1, or RQI (RNA Qualified; Promega) DNase was for 15 min at 37°C.
For degradation assays, templates specifying the transcripts, as shown in Figure 5A, were amplified using PCR, and in vitro transcription was performed using T7 RNA polymerase with the inclusion of -32P-ATP. These transcripts were gel purified and polyadenylated in vitro using yeast poly(A) polymerase, under the conditions specified by the manufacturer. Under these conditions, partial polyadenylation was obtained, and the resulting mixture of polyadenylated and nonpolyadenylated transcripts was incubated at 37°C with the S100 extract (450 µg of protein per reaction) in the presence of poly(A) polymerase buffer and 1 x MTB (Rapp and Stern 1992 ). Reactions were stopped as described for poly(A) addition.
PCR Amplification and Sequencing to Determine the Editing Status of Bulk mtRNA
A 30-µL reaction containing 90 ng of DNase-treated mtRNA was reverse transcribed by using 450 ng of the cox2-specific oligonucleotide ZMC24, extending from nucleotides 1762 to 1748 of the cox2 sequence, and avian myeloblastosis virus reverse transcriptase (Promega) for 1 hr at 42°C. Then, 2 µL of this product was PCR amplified by using oligonucleotides ZMC24 (see above) and ZMC25, which extends from nucleotides 1293 to 1305 of the cox2 sequence. Amplification was performed for 25 cycles of 1 min each at 94, 50, and 72°C. The product was isolated in low-melting-temperature agarose and directly sequenced as originally described (Bachmann et al. 1990 ) in a reaction containing manganese (to read the sequences close to the primer) using ZMC26, which extends from nucleotide 1314 to 1329 of the cox2 sequence. The bands in Figure 6 representing the edited (lanes T) and unedited (lanes C) bases for each edited position (as well as nearby unedited T and C bands as controls for variation in band intensity related to sequence length and composition) were quantified using a PhosphorImager (Molecular Dynamics, Sunnyvale, CA). The values given represent the averages of three experiments.
|
|
ACKNOWLEDGMENTS |
|---|
We thank Irena Lisitsky for contributing to the inception of this project and Gadi Schuster for numerous helpful comments and advice. We thank members of the Stern laboratory for helpful discussions. D.S.L. was supported by a National Institutes of Health (NIH) training grant to the Section of Genetics and Development. This work was supported by NIH Grant No. R01GM52560 to D.B.S.
|
|
REFERENCES |
|---|
Avadhani, N.G. (1979) Messenger ribonucleic acid metabolism in mammalian mitochondria: Relationship between the decay of mitochondrial mRNA and their poly(A). Biochemistry 18:2673-2678[CrossRef][Medline].
Bachmann, B., Luke, W., and Hunsmann, G. (1990) Improvement of PCR amplified DNA sequencing with the aid of detergents. Nucleic Acids Res. 18:1309[Free Full Text].
Bariola, P.A., Howard, C.J., Taylor, C.B., Verburg, M.T., Jaglan, V.D., and Green, P.J. (1994) The Arabidopsis ribonuclease gene RNS1 is tightly controlled in response to phosphate limitation. Plant J. 6:673-685[CrossRef][Web of Science][Medline].
Bellaoui, M., Pelletier, G., and Budar, F. (1997) The steady-state level of mRNA from the Ogura cytoplasmic male sterility locus in Brassica cybrids is determined post-transcriptionally by its 3' region. EMBO J. 16:5057-5068[CrossRef][Web of Science][Medline].
Bhat, G.J., Souza, A.E., Feagin, J.E., and Stuart, K. (1992) Transcript-specific developmental regulation of polyadenylation in Trypanosoma brucei mitochondria. Mol. Biochem. Parasitol. 52:231-240[CrossRef][Medline].
Brewer, G., and Ross, J. (1988) Poly(A) shortening and degradation of the 3' A+U-rich sequences of human c-myc mRNA in a cell-free system. Mol. Cell. Biol. 8:1697-1708[Abstract/Free Full Text].
Cermakian, N., Ikeda, T.M., Cedergren, R., and Gray, M.W. (1996) Sequences homologous to yeast mitochondrial and bacteriophage T3 and T7 RNA polymerases are widespread throughout the eukaryotic lineage. Nucleic Acids Res. 24:648-654[Abstract/Free Full Text].
Chang, C., Sheen, J., Bligny, M., Niwa, Y., Lerbs-Mache, S., and Stern, D.B. (1999) Functional analysis of two maize cDNAs encoding T7-like RNA polymerases. Plant Cell 11:911-926[Abstract/Free Full Text].
Dombrowski, S., Brennicke, A., and Binder, S. (1997) 3'-Inverted repeats in plant mitochondrial mRNAs are processing signals rather than transcription terminators. EMBO J. 16:5069-5076[CrossRef][Web of Science][Medline].
Fauron, C.M.R., and Havlik, M. (1988) The BamHI, XhoI, and SmaI restriction enzyme maps of the normal maize mitochondrial genome genotype B37. Nucleic Acids Res. 16:10395-10396[Free Full Text].
Feagin, J.E., Jasmer, D.P., and Stuart, K. (1985) Apocytochrome b and other mitochondrial DNA sequences are differentially expressed during the life cycle of Trypanosoma brucei.. Nucleic Acids Res. 13:4577-4596[Abstract/Free Full Text].
Finnegan, P.M., and Brown, G.G. (1990) Transcriptional and post-transcriptional regulation of RNA levels in maize mitochondria. Plant Cell 2:71-84[Abstract/Free Full Text].
Fox, T.D., and Leaver, C.J. (1981) The Zea mays mitochondrial gene coding cytochrome oxidase subunit II has an intervening sequence and does not contain TGA codons. Cell 26:315-323[CrossRef][Web of Science][Medline].
Gagliardi, D., and Leaver, C.J. (1999) Polyadenylation accelerates the degradation of the mitochondrial mRNA associated with cytoplasmic male sterility in sunflower. EMBO J. in press.
Gelfand, R., and Attardi, G. (1981) Synthesis and turnover of mitochondrial ribonucleic acid in HeLa cells: The mature ribosomal and messenger ribonucleic acid species are metabolically unstable. Mol. Cell. Biol. 1:497-511[Abstract/Free Full Text].
Gray, M.W. (1992) The endosymbiont hypothesis revisited. Int. Rev. Cytol. 141:233-357[Web of Science][Medline].
Groot, G.S., Flavell, R.A., Van Ommen, G.J., and Grivell, L.A. (1974) Yeast mitochondrial RNA does not contain poly(A). Nature 252:167-169[CrossRef][Medline].
Hajnsdorf, E., Braun, F., Haugel-Nielsen, J., Le Derout, J., and Regnier, P. (1996) Multiple degradation pathways of the rpsO mRNA of Escherichia coli. RNase E interacts with the 5' and 3' extremities of the primary transcript. Biochimie 78:416-424[Medline].
Hanic-Joyce, P.J., and Gray, M.W. (1990) Processing of transfer RNA precursors in a wheat mitochondrial extract. J. Biol. Chem. 265:13782-13791[Abstract/Free Full Text].
Haugel-Nielsen, J., Hajnsdorf, E., and Regnier, P. (1996) The rpsO mRNA of Escherichia coli is polyadenylated at multiple sites resulting from endonucleolytic processing and exonucleolytic degradation. EMBO J. 15:3144-3152[Web of Science][Medline].
Hayes, R., Kudla, J., and Gruissem, W. (1999) Degrading chloroplast mRNA: The role of polyadenylation. Trends Biochem. Sci. 24:199-202[CrossRef][Web of Science][Medline].
Hedtke, B., Borner, T., and Weihe, A. (1997) Mitochondrial and chloroplast phage-type RNA polymerases in Arabidopsis. Science 277:809-811[Abstract/Free Full Text].
Hendler, F.J., Padmanaban, G., Patzer, J., Ryan, R., and Rabinowitz, M. (1975) Yeast mitochondrial RNA contains a short polyadenylic acid segment. Nature 258:357-359[CrossRef][Medline].
Hirsch, M., and Penman, S. (1973) Mitochondrial polyadenylic acid–containing RNA: Localization and characterization. J. Mol. Biol. 80:379-391[CrossRef][Medline].
Kaleikau, E.K., Andre, C.P., and Walbot, V. (1992) Structure and expression of the rice mitochondrial apocytochrome b gene cob-1 and pseudogene cob-2. Curr. Genet. 22:463-470[CrossRef][Web of Science][Medline].
Kempken, F., and Howad, W. (1996) Mitochondrial RNA editing is sequence specific and independent of transcript abundance in Sorghum bicolor. Curr. Genet. 30:186-189[Medline].
Koslowsky, D.J., and Yahampath, G. (1997) Mitochondrial mRNA 3' cleavage/polyadenylation and RNA editing in Trypanosoma brucei are independent events. Mol. Biochem. Parasitol. 90:81-94[CrossRef][Web of Science][Medline].
Kudla, J., Hayes, R., and Gruissem, W. (1996) Polyadenylation accelerates degradation of chloroplast mRNA. EMBO J. 15:7137-7146[Web of Science][Medline].
Li, Q., and Hunt, A.G. (1997) The polyadenylation of RNA in plants. Plant Physiol. 115:321-325[CrossRef][Web of Science][Medline].
Li, Z., Pandit, S., and Deutscher, M.P. (1998) Polyadenylation of stable RNA precursors in vivo. Proc. Natl. Acad. Sci. USA 95:12158-12162[Abstract/Free Full Text].
Lisitsky, I., Klaff, P., and Schuster, G. (1996) Addition of poly(A)-rich sequences to endonucleolytic cleavage sites in the degradation of spinach chloroplast mRNA. Proc. Natl. Acad. Sci. USA 93:13398-13403[Abstract/Free Full Text].
Lisitsky, I., Kotler, A., and Schuster, G. (1997) The mechanism of preferential degradation of polyadenylated RNA in the chloroplast: The exoribonuclease 100RNP-polynucleotide phosphorylase displays high binding affinity for poly(A) sequence. J. Biol. Chem. 272:17648-17653[Abstract/Free Full Text].
Lonsdale, D.M., Hodge, T.P., and Fauron, C.M.-R. (1984) The physical map and organization of the mitochondrial genome from the fertile cytoplasm of maize. Nucleic Acids Res. 12:9249-9261[Abstract/Free Full Text].
Lu, B., and Hanson, M.R. (1992) A single nuclear gene specifies the abundance and extent of RNA editing of a plant mitochondrialtranscript. Nucleic Acids Res. 20:5699-5703[Abstract/Free Full Text].
Maier, R.M., Zeltz, P., Kossel, H., Bonnard, G., Gualberto, J.M., and Grienenberger, J.M. (1996) RNA editing in plant mitochondria and chloroplasts. Plant Mol. Biol. 32:343-365[CrossRef][Web of Science][Medline].
Marchfelder, A., and Brennicke, A. (1994) Characterization and partial purification of tRNA processing activities from potato mitochondria. Plant Physiol. 105:1247-1254[Abstract].
Moore, A.L., and Proudlove, M.O. (1983) Mitochondria and sub-mitochondrial particles. In Moore A.L., ed. Isolation of Membranes and Organelles from Plant Cells. London, Academic Press. 153–184.pp.
Mulligan, R.M., Leon, P., and Walbot, V. (1991) Transcriptional and post-transcriptional regulation of maize mitochondrial gene expression. Mol. Cell. Biol. 11:533-543[Abstract/Free Full Text].
Ojala, D., Montoya, J., and Attardi, G. (1981) tRNA punctuation model of RNA processing in human mitochondria. Nature 290:470-474[CrossRef][Medline].
Pereira de Souza, A., Jubier, M.F., Delcher, E., Lancelin, D., and Lejeune, B. (1991) A trans-splicing model for the expression of the tripartite nad5 gene in wheat and maize mitochondria. Plant Cell 3:1363-1378[Abstract/Free Full Text].
Rapp, W.D., and Stern, D.B. (1992) A conserved 11 nucleotide sequence contains an essential promoter element of the maize mitochondrial atp1 gene. EMBO J. 11:1065-1073[Web of Science][Medline].
Rose, K.M., Morris, H.P., and Jacob, S.T. (1975) Mitochondrial poly(A) polymerase from a poorly differentiated hepatoma: Purification and characteristics. Biochemistry 14:1025-1032[CrossRef][Medline].
Rothnie, H.M. (1996) Plant mRNA 3'-end formation. Plant Mol. Biol. 32:43-61[CrossRef][Web of Science][Medline].
Sambrook, J., Fritsch, E.F., and Maniatis, T. (1989) Molecular Cloning: A Laboratory Manual, 2nd ed, Vol. 2. Cold Spring Harbor, NY, Cold Spring Harbor Laboratory Press.
Sarkar, N. (1997) Polyadenylation of mRNA in prokaryotes. Annu. Rev. Biochem. 66:173-197[CrossRef][Web of Science][Medline].
Stern, D.B., and Newton, K.J. (1986) Isolation of plant mitochondrial RNA. Methods Enzymol. 118:488-496[Medline].
Stern, D.B., Dyer, T.A., and Lonsdale, D.M. (1982) Organization of the mitochondrial ribosomal RNA genes of maize. Nucleic Acids Res. 10:3333-3340[Abstract/Free Full Text].
Tracy, R.L., and Stern, D.B. (1995) Mitochondrial transcription initiation: Promoter structures and RNA polymerases. Curr. Genet. 28:205-216[CrossRef][Web of Science][Medline].
Wahle, E., and Keller, W. (1992) The biochemistry of 3'-end cleavage and polyadenylation of messenger RNA precursors. Annu. Rev. Biochem. 61:419-440[Web of Science][Medline].
Wilson, R.K., and Hanson, M.R. (1996) Preferential RNA editing at specific sites within transcripts of two plant mitochondrial genes does not depend on transcriptional context or nuclear genotype. Curr. Genet. 30:502-508[CrossRef][Web of Science][Medline].
Xu, F., Lin Chao, S., and Cohen, S.N. (1993) The Escherichia coli pcnB gene promotes adenylation of antisense RNAI of ColE1-type plasmids in vivo and degradation of RNAI decay intermediates. Proc. Natl. Acad. Sci. USA 90:6756-6760[Abstract/Free Full Text].
Yang, A.J., and Mulligan, R.M. (1991) RNA editing intermediates of cox2 transcripts in maize mitochondria. Mol. Cell. Biol. 11:4278-4281[Abstract/Free Full Text].
Yen, Y., and Green, P.J. (1991) Identification and properties of the major ribonucleases of Arabidopsis thaliana. Plant Physiol. 97:1487-1493[Abstract/Free Full Text].
Young, D.A., Allen, R.L., Harvey, A.J., and Lonsdale, D.M. (1998) Characterization of a gene encoding a single-subunit bacteriophage-type RNA polymerase from maize that is alternatively spliced. Mol. Gen. Genet. 260:30-37[CrossRef][Web of Science][Medline].
Yuckenberg, P.D., and Phillips, S.L. (1982) Oligoadenylate is present in the mitochondrial RNA of Saccharomyces cerevisiae. Mol. Cell. Biol. 2:450-456[Abstract/Free Full Text].
TOP
ABSTRACT
INTRODUCTION
