The Plant Cell, Vol. 3, 1109-1120, October 1991 O 1991 American Society of Plant Physiologists

Expression of the Wheat Mitochondrial nad3-rpsl2 Transcription Unit: Correlation between Editing and mRNA Maturation

Jose Manuel Gualberto, Geraldine Bonnard, Lorenzo Lamattina, and Jean Michel Grienenberger' lnstitut de Biologie Moléculaire des Plantes du Centre National de Ia Recherche Scientifique, Université Louis Pasteur, 12 rue du Général Zimmer, 67084 Strasbourg Cedex, France

In plant mitochondria, RNA editing involves the conversion of cytidines in the genomic DNA into uridines in the corresponding RNA. Analysis of cDNAs prepared by reverse transcription of mitochondrial RNAs has shown that partially edited RNAs are present in wheat mitochondria. The extent of this partial editing as well as its potential influence on the corresponding protein sequence were studied along with the expression of a wheat mitochondrial locus. The sequence, expression, and RNA editing of the wheat mitochondrial transcription unit containing four open reading frames (nad3, rps72, orf299, and orf756), all cotranscribed into a same predominant precursor RNA, have been studied. The product of orf756 is an 18-kD mitochondrial membrane protein of unknown function, whereas the product of orf299 could not be detected and this sequence seems to be a pseudogene. Sequences of cDNA clones derived by the polymerase chain reaction technique show that nad3, rps72, and otf756 transcripts are edited, whereas orf299 is not edited, except for a sequence identical to part of the coxll gene. Analysis of cDNA clones obtained from the precursor RNA shows the presence of a large number of partially edited nad3-rps72 transcripts with no evident polarity for the editing process. This shows that RNA editing is a post-transcriptional event. In addition, study of partial editing at the leve1 of precursor, mature, and polysomal transcripts shows that mainly mature, completely edited sequences are used for translation. Deletions of a nucleotide at editing sites were observed in a number of cDNA clones, suggesting that C 4 J RNA editing in plant mitochondria would be achieved by nucleotide replacement.

INTRODUCTION

The factors involved in the regulation of gene expression in plant mitochondria are still largely unknown. Differential gene expression can be regulated at the levels of tran- scription initiation and RNA stability. Run-on transcription experiments have shown that both transcriptional and post-transcriptional regulation are required to maintain transcript steady-state levels in plant mitochondria (Finnegan and Brown, 1990; Mulligan et al., 1991). Fur- thermore, the cotranscription of genes coding for proteins, which are involved in different metabolic pathways (Bland et al., 1986; Gualberto et al., 1988; Wissinger et al., 1988), suggests that translational regulation also plays an impor- tant role.

The recently described CAU RNA editing activity in plant mitochondria (Covello and Gray, 1989; Gualberto et al., 1989; Hiesel et al., 1989; Lamattina et al., 1989) may have an important role in the control of translation: all plant mitochondrial protein genes analyzed so far are more or

' To whom correspondence should be addressed.

less affected by editing, resulting in significant modifica- tions of the predicted protein sequence. The edited tran- scripts are substrates for translation as confirmed by the sequence of the wheat ATP9 protein (Bégu et ai., 1990). However, no plant mitochondrial tRNA or rRNA has been shown to be edited, therefore suggesting that editing is required only for the expression of plant mitochondrial protein genes. Up to now, nothing is known about either the enzymatic activities required for editing or the possible function of editing on plant mitochondrial gene expression. One may wonder whether the plant mitochondrial editing system has common characteristics with other mitochon- drial editing systems, such as those in trypanosomes and Physarum polycephalum (Benne et al., 1986; Mahendran et al., 1991). Although these different editing systems result in different RNA sequence modifications, the broad principles of mitochondrial editing may have been con- served during evolution, as well as the functional role that editing may have on mitochondrial gene expression.

The effect of RNA editing on the expression of the wheat mitochondrial transcription unit that contains the nad3 and

11 1 O The Plant Cell

Lz N2

I 500 bp ,

0.9 kb 1.3 kb

1.1 kb

0.8 kb ...

3.0 kb

Figure 1. Structure of the Wheat nad3-rpsl2 Transcription Unit.

The ORFs are indicated by open boxes. The dotted box represents the coxll homologous sequence. A tRNA gene and a tRNA pseudogene present upstream of nad3 (Gualberto et al., 1988) are indicated by dark boxes. Arrows indicate positions mapped as transcript extremities, whereas the black bars indicate DNA regions corresponding to the 3.0-, 1.3-, 1 .l-, 0.9-, and 0.8-kb transcripts. Relevant restriction sites are labeled as follows: E, EcoRI; S, Sall; Sc, Scal; Sm, Smal; X, Xhol. Restriction fragments used as probes on the RNA gel blots of Figure 3 are numbered 1 to 5.

rps72 genes (coding for subunit 3 of NADH dehydrogenase and for ribosomal protein S12, respectively) and two uni- dentified open reading frames (orf299 and orf756) has been studied. RNA and protein gel blot analyses together with cDNA sequence determination gave evidence that editing in plant mitochondria is sequence specific and does not require translation, although only translatable se- quences were found to be edited. Partia1 editing of nad3-rps72 was studied for precursor, mature, and poly- soma1 transcripts. A correlation between RNA editing and RNA maturation has been found, and polysomal tran- scripts seem to be edited to a higher extent than tran- scripts from total RNA. cDNA clones were found where a nucleotide has been deleted at an editing position. This observation suggests that editing is achieved by base replacement, with or without cleavage of the RNA.

RESULTS

Sequence and Transcription of the nad3-rps72 Transcription Unit

The nad3 and rps72 genes are present in the wheat mitochondrial genome in the Sal1 fragment L2 (Quétier et al., 1985) and are cotranscribed into a precursor transcript of 3.0 kb extending into the contiguous N2 fragment (Gualberto et al., 1988). The corresponding DNA regions have been sequenced, revealing two additional open read- ing frames (ORFs) of 299 codons (orf299) and 156 codons

(orf756) that are present downstream of rps72, as shown in Figures 1 and 2. Both ORFs code for putative proteins that could not be identified by their homology with other known proteins. The sequence of orf299 contains a region of 193 bp that is a duplication of part of the first exon of the wheat gene coding for subunit II of cytochrome oxidase (coxll). The presence of this repeated sequence in the wheat mitochondrial genome had already been reported (Bonen et al., 1984) and shown to correspond to the transmembrane part of the COXll protein. Therefore, it would be possible that if orf299 were expressed, the 64 amino acids coded by the coxll sequence would corre- spond to a membrane-spanning domain. On the other hand, homologous sequences could not be found by hy- bridization in the maize mitochondrial genome, indicating that the orf299 sequence has not been conserved during evolution.

The predicted product of orf156 shows a strong similar- ity (86%) with the predicted product of Oenothera mito- chondrial orfB, a sequence cotranscribed in Oenothera and sunflower with the coxlll gene (Hiesel et al., 1987; Quagliariello et al., 1990). An homologous sequence could also be identified by hybridization in the maize mitochon- drial genome but at a different locus than the nad3, rps72, and coxlll genes (not shown).

Transcription of the different ORFs was studied by RNA gel blot hybridization, as shown in Figure 3. Probes used are indicated in Figure 1 and numbered 1 to 5. The 0.9-kb mRNA of nad3 and rps72 exists in wheat mitochondria at similar (see RNA gel blot hybridizations in Gualberto et al., 1988) or slightly higher amounts than the precursor 3.0-kb RNA (Figure 3). The estimated ratios between the

RNA Editing and Maturation in Wheat Mitochondria 1111

nad3 M L E F A P L G V P F L F A SS L L V S L

N S S T Y P E K L S A Y E C G F D P F G D A R S R F D R F Y L V S

D L E V T F F F P W A V S L N K I L F G F W S M M A F L L I L T I G F L Y E W

I R G V K D L L G I P D R R K G R S K Y G A E R P K S K

CAGTGCAGGAAATTGTGGTCCTCAAAACAATGTGTCCC(j\Tli\TCATTtrTACATCACGATATCTTTTTCTTCCTCATTCTTATTTTGGTTTTCGTATfAfCGATGTTGCTTCCCGCTTTA

^AAAAGGGGTGCTTTGGATTGGGAGTAACCACTTTAGAAAGGGCAAAGGGGGGAACGACATAGGAAAGAGGGATGCCTACAAAAAATCAATTGATTCGTCATGGTAGAGAAGAAAAACGGC 480K R G A L O W E 4 rpslZ M P T K N Q L I R H G R E E K R R J 7

T D R T R A L D Q C P Q K Q G V C L R V S T R T P K K P N S A L R K I A K V R L 5 7• • • • •S N R H D I F A Y I P G E G H N L Q E H S I V L V R G G R V K D L P G V K F H C 9 7

84002 125

960TGACAGGGCCGCCCCGGCAGCGTGGCTTCTCTGCCGAGCAGGAAGCCACAAAAACCATTGTGCTCTTGTAAetgatettttaota^fictaaqttGCTGAATTCTGAAAT^ACTACAGGCA 1080

03 04 _______________________

~" " 120013201440156016801800192020402160

102280SO240090252013026401562760

[£AC|GTTGGCCGCCAACTTCGCCTG^G'TTTCTATCTGAGTTCTTCicdTCTTCATCCTTTCGAACGAC'fceirAAATTTCACAAAA^gTTTTTTTCTTATTTGAAATCCAAATCCAAATr.CCTCAACTTGATAAATTAACTTATTT

0 5 orf!56 M P Q L D K L T Y F• Ub •CTCACAATTCTTCTGGTTATGTCTTCTCTTCTTTACTTTTTATATTCTCTTATTTAATAATAATAATGCAATACTTCCAATTAGTAGAATTCTTAAACTACCCAACCAACTGCTTTCGCA

S Q F F W L C L L F F T F Y I L L F N N N N G I L G I S R I L K L R N Q L L S H

R G G E I R S K D P K N L E D I L R K G F S T G L S Y M Y S S L S E V S Q W C K

T V D Y L G K R R K 1 T L I S D F G E I S G S R G M E R O I L Y L I S K S S Y N

T S S S R I T C W K N I M L T H V L H G Q G S I I S *AATGGAAAAATCACAAAAACACTTTCTATATGTTCTCTTATTTCGAGATTCGAAGAAGCAACCAAGTTtttcatatafictatttcctatntantqatctctattcatnaanateataaoo

07cccctnataccctcj tqatacnngggfj t j^^taqaaijaatQnncatgtqqgcttctSatr taaaaqaq^aaaa^atacnqntat-rqctatqctratat ir t r-cct tccctcctccaqgt^aaaaoctggtgt t gagaaatqacatccaatctt t tqqtatcqatcatatat tctaggtgqatcact t t t t t tc t t caactqaaatqqgctgatctagtgt t t t t aaatgaccttccccgatcagagaagggcaagaactctcttgaacagtgaggagcacaacgatttgtgctctgagcgatacagcttctgtaaaggagtacaaggtgctgtgctaataaaaggagaaatgagataaggagcagt aaggagqataaaagaagtagagt taggttct tgtgtgtaatcgatat gtctcccgggagaaaaccctct atcaacacaactatccgaacacctatagctagcttgctttagctccaact

2880300031203240

Figure 2. Edited Sequence of the Transcription Unit.

The sequence is the consensus derived from the cDNA clones analyzed. The protein sequences presented are the ones derived fromcompletely edited cDNA. Edited nucleotides (corresponding to a genomic C converted into a T in the cDNA sequence) are indicated bydots. Regions whose cDNA sequence has not been determined and that corresponds to the genomic sequence are in lower case.Sequences corresponding to oligonucleotides O1 to O7 utilized in cDNA amplification by PCR are underlined. The initiation and terminationcodons of orf299 and the cox// homologous sequence are boxed with a continuous line. Nucleotides mapped as transcript extremitiesare boxed with a dotted line. Sequences that can base pair in a characteristic 3' end double-loop structure are underlined with arrows.

kb

1 2 3 4 5

SW-" A B237 238*- 4 -

•I1.3 170 171 »>

Zr131r™uf'-it;rf

Figure 3. Transcription Analysis.(A) RNA gel blots of total wheat mtRNA hybridized with probesindicated in Figure 1. Transcript lengths were determined usingRNA length markers (Bethesda Research Laboratories).

two transcripts can change using RNA from different ex-tractions, suggesting that the 3.0-kb RNA is less stablethan the 0.9-kb RNA. Both orf299 and orf156 are tran-scribed into the same 3.0-kb precursor, but a number ofadditional transcripts, of lengths spanning the range from1.3 to 0.8 kb, have been identified by hybridization withprobes 3,4, and 5. These RNAs are likely to be processing

(B) Primer extension analysis of the 5' end of orf156 transcripts.The lengths of the cDNA fragments obtained by extension ofprimer O6 are indicated in nucleotides. The ladder is the DNAsequence primed by the same oligonucleotide.(C) S1 nuclease mapping of the transcription unit 3' end. Thelengths of the protected fragments are in nucleotides and weredetermined using the M13mp18 sequence as a ladder.

1112 The Plant Cell

products of the 3.0-kb RNA, but the hypothesis that alter-native sites of transcription initiation exist cannot be ex-cluded. The transcripts of about 0.8 kb exist in highamounts and are specific for orf156. As revealed by shortertimes of autoradiography (not shown), the 0.8-kb band isbroad and could correspond to RNAs of slightly differentlengths. Less abundant transcripts of 1.3 and 1.1 kbcontain the orf156 sequence with parts of the orf299sequence. Probe 2 (Figure 1) hybridized to a 1.3-kb bandcorresponding to the cox// mRNA (Bonen et al., 1984) andto a transcript of the same length that could be specificallyidentified by hybridization with probe 3 (Figure 3A). How-ever, no transcript specific for orf299 could be identified,either with the indicated probes or with others. The onlyRNA that contains the entire orf299 sequence is, therefore,the 3.0-kb precursor.

Termini of the different transcripts were mapped. Thenad3-rps12 0.9-kb transcript had been previously charac-terized (Gualberto et al., 1988). The 5' termini of the RNAspecies specific for orf156 were mapped by primer exten-sion, using oligonucleotide O6 (Figure 2) as primer forcDNA synthesis. Five predominant signals were revealedon autoradiograms (Figure 3B), corresponding to nucleo-tides 30 to 115 bp upstream of orf156. Other weak signalswould correspond to less abundant transcripts or to pre-mature termination of elongation products. The 3' end ofthe transcription unit was mapped by S1 nuclease protec-tion utilizing a 765-bp Hinfl fragment as a probe. Protectedfragments 128 to 131 nucleotides long (Figure 3C) mappeda 3' extremity, which is common to all identified transcriptsexcept for the nad3-rps12 0.9-kb RNA. The 3' end corre-sponds to a sequence predicted to fold into a double stem-loop structure (Figure 2). Homologous structures havebeen described in the 3' ends of other plant mitochondrialtranscripts (Schuster et al., 1986).

fraction of the bacteria and were further purified by pre-parative SDS-PAGE before being injected into rabbits toraise polyclonal antisera. In protein gel blots of wheatmitochondrial protein extracts, sera specific for the fusionprotein expressed by clone pEA305:ORF156 specificallyrecognize a polypeptide associated with the mitochondrialmembrane fraction, as shown in Figure 4. The estimatedsize of the polypeptide (18 kD) corresponds to the sizeexpected for the product of orf156. Sera against the fusionprotein expressed by clone pEA305:ORF299 cannot rec-ognize any wheat mitochondrial protein at a level higherthan background signals (not shown). The same sera can,however, specifically recognize the corresponding fusionprotein. It seems, therefore, that orf299 is not expressedin wheat mitochondria under these physiological conditionsand is possibly a pseudogene.

RNA Editing of the or/299 and or/756 Transcripts

Amplified cDNA clones of orf299 were obtained utilizingprimers 04 and 06. The 1202-bp amplified cDNA was cutwith EcoRI (at position 1715), and the resulting 0.65- and0.5-kb fragments were independently cloned and se-quenced. Five clones of the 0.65-kb fragment and fourclones of the 0.5-kb fragment were sequenced. Only threeediting positions (C—>U conversions) could be found in theentire sequence of orf299, all three in the sequence cor-responding to the cox// insertion (Figure 2). These edited

1 2 3

Expression of or/299 and or/756

To determine whether the protein products encoded byorf299 and orf156 are present in wheat mitochondria,fusion proteins were expressed in Escherichia coli, andspecific antibodies were obtained. A 989-bp Scal-EcoRVfragment (from coordinates 1338 to 2326 of the sequence)containing the last 226 codons of orf299 and a 928-bpEcoRI-Smal fragment (from coordinates 2248 to 3175 ofthe sequence) containing the last 117 codons of orf156have been cloned in the Hindlll site of the expressionvector pEA305AHindlll (Amann et al., 1983; Schmidt etal., 1986). This vector allows the expression of fusionproteins with the first 163 amino acids of the represserprotein cl of A. This expression is under the control of theinducible fac promoter. The clones obtained were namedpEA305:ORF299 and pEA305:ORF156, respectively.After induction by isopropyl /3-o-thiogalactoside, theexpressed fusion proteins accumulated in the insoluble

18 kD

Figure 4. Immunodetection of the orf156 Product.

Wheat mitochondrial proteins transferred to nitrocellulose mem-branes were probed with specific sera against the cl-ORF156fusion protein. Lane 1, total mitochondrial protein; lane 2, mem-brane fraction; lane 3, soluble fraction; lane 4, proteins frommembrane fraction solubilized with 1 % sodium deoxycholate.

RNA Editing and Maturation in Wheat Mitochondria 11 13

nucleotides correspond to those edited in the homologous sequence of the wheat coxll mRNA (Covello and Gray, 1989; Gualberto et al., 1989). The 727-bp sequence unique to orf299 is not edited.

Amplified cDNA clones of orf756 were obtained with; primers 0 5 and 07. All six independent clones that were sequenced contained the same four edited positions ( C - 4 conversions), corresponding to 3 amino acid sub- stitutions and to one silent modification (Figure 2).

RNA Editing of the nad3 and rpsl2 Transcripts

The positions where editing occurs in the transcripts of the nad3-rps 72 transcription unit have been determined by sequencing cDNA clones derived by the polymerase chain reaction (PCR) technique, as described previously for the wheat coxlll mRNA (Gualberto et al., 199Ob). cDNA clones of nad3 and rps72 were obtained utilizing primers 01, 02, and 0 3 (Figure 2). Because the 3' primer 0 3 is complementary to sequences absent from the mature RNA of 0.9 kb, PCR amplification with this primer specifically yields cDNAs from the 3.0-kb transcript. Amplification with 3' primer 0 2 yields cDNAs that could come either from the 0.9- or the 3.0-kb RNA. Only modifications to the

genomic sequence found in more than three different clones were considered as resulting from RNA editing. Respectively, 21 and six editing sites could be identified in the nad3 and rps72 sequences, all corresponding to C 4 J conversions (Figure 2). When the amino acids spec- ified by these edited codons are analyzed, 16 amino acids of nad3 and 6 amino acids of rps72 are modified. In nad3 mRNA, editing of site 2 (codon 13) is silent and codons 15, 21, 27, and 70 contain two editing sites.

Partia1 Editing of the nad3 and rpsl2 Transcripts

The existence of partially edited transcripts has already been postulated based on sequencing of uncloned nad3 cDNAs (Gualberto et al., 1989). The sequencing of a large number of nad3-rps72 cDNA clones confirms this obser- vation: a significant fraction of the sequenced clones cor- responds to partially edited or unedited transcripts. As a consequence, proteins translated from either completely edited or partially edited RNAs would have sequences differing at some amino acid positions. Many of these positions, as inferred by homology with the corresponding proteins from nonplant organisms, must be essential for protein function.

3 5 13 1 2 3 4 5 6

number of + + + + # clones C-U H s 12

2 3 1 1 1 1 1 1 1 1 1 i 1 i 1 7

1 1 1 1 1 1 1 1 1 1 5

20/6 18/6 18/6 18/6 20 /3 20/3 i5/5 16/2 18/0 1/5 2/2 0 / 3 1/1 1/0 0/1 o/o

20 19 16 15 11

6 3 1 1 1 O

m pz E3 w €5

Pi W €9 P10 Pll Plí P13 P14 P15 P16

F28 F29 P30 P31 P32 P33 P42 P35 P36 P37 P38

m

Figure 5. Editing of nad3 and rps72 Sequences in the 3.0-kb Transcript.

Represented here are the sequences of cDNA clones specifically obtained from the 3.0-kb transcript of total mtRNA. The relative positions of the 21 and six editing sites of nad3 and rps72, respectively, are represented by vertical arrows. Each line corresponds to a sequence found in one or more cDNA clones (number of clones indicated). Dots represent nucleotides found edited in the cDNA sequences. The number of nucleotides edited in each cDNA sequence is indicated. Sequences P1 to P16 correspond to cDNA clones containing both nad3 and rps72 sequences, whereas sequences P17 to P38 correspond to cDNA clones either containing nad3 or rps72.

11 14 The Plant Cell

The hypothesis that only completely edited RNA se- quences are translated in mitochondria was considered. The existence of partially edited RNA could be a conse- quence of the accumulation of nontranslatable precursor transcripts in plant mitochondria. Considering this hypoth- esis, we would expect to find precursor transcripts of nad3 and rps72 less edited than mature transcripts and tran- scripts isolated from mitochondrial polysomes completely edited. To test this hypothesis, partia1 editing of nad3 and rps72 was studied at the leve1 of precursor, mature, and polysomal mitochondrial RNA (mtRNA) fractions, and the corresponding data are shown in Figures 5, 6, and 7.

Mature nad3 and rps l2 Transcripts Are More Edited than Precursor Transcripts

Specific nad3 and rps72 cDNA clones from the precursor 3.0-kb RNA or from the mature 0.9-kb RNA were obtained, and from each cDNA population a large number of PCR- derived clones were sequenced. Specific clones from the 3.0-kb RNA were obtained with primers O1 and 0 3 (Figure 2). To compare editing of the two genes, clones of the 1031 -bp, PCR-derived cDNA were completely sequenced. Each type of clone is represented in P1 to P16 of Figure 5. Clones containing either the nad3 or the rpsl2 se- quences were obtained by cutting the PCR fragment with Xhol (at position 494 of the sequence) and separately cloning the resulting fragments (clones P17 to P38 of Figure 5). Specific cDNA clones from the mature 0.9-kb RNA were obtained by size fractionation of total mtRNA

on sucrose gradients. Fractions containing only the mature RNA of 0.9 kb were selected by RNA gel blot hybridization and utilized for cDNA synthesis and amplification with primers O1 and 0 2 (Figure 2).

Most of the 3.0-kb RNA clones corresponded to non- edited and partially edited transcripts (Figure 5). In con- trast, Figure 6 shows that most of the cDNA clones from the 0.9-kb mature RNA are completely or almost com- pletely edited. None of the 37 cDNA clones is unedited, and although some clones (clones M16, M17, M20, and M21) are unedited in one of the genes, the other gene is edited to some extent. As shown in Figures 8A and 8B, most of the 0.9-kb RNA clones code for the final NAD3 and RPS12 proteins, whereas a large proportion of the 3.0-kb RNA clones are still not edited. Although most 0.9-kb RNA clones are completely edited, a significant number are partially edited, potentially coding for NAD3 and RPS12 proteins of altered sequence. Many of these clones contain nearly completely edited nad3 sequences, possibly corresponding to transcripts in final steps of the editing process.

We could not find any polarity to the sequence of editing events (neither from 5‘ to 3’ nor from 3’ to 5’) and no precise site of editing initiation. The sequences of some clones are, however, consistent with the idea that although nad3 and rps72 are cotranscribed, editing of each gene sequence is independently regulated: clones P9, M16, and M17 have completely or almost completely edited nad3 sequences and nonedited rps72 sequences, but clone P1 O has only one edited site on nad3 and a nearly completely edited rps72 sequence.

3 21/6 13 20/6 3 19/6 1 19/6 1 19/6 1 19/6 1 19/5 1 18/6 1 18/6 1 19/5 1 18/5 1 17/6 1 17/6 1 20/2 1 20/1 1 20/0 1 16/0 1 5/4 1 2/2 1 0 /3 1 0/1

Figure 6. Editing of nad3 and rps72 Sequences on the Transcript of 0.9 kb.

Shown here are the sequences of cDNA clones specifically obtained from the 0.9-kb transcript of total rntRNA. The notations are the sarne as given in Figure 5. Open circles indicate positions where a nucleotide at an editing site has been deleted in individual cDNA clones (see Results and Figure 9).

RNA Editing and Maturation in Wheat Mitochondria 11 15

3 s 13 2 4 6 78 11 121415 1617 18 19 2021 I 2 3 4 5 6

number of t t t i # I-

1 16/6 1 15/6 1 14/6 1 8/6 1 6/6 1 5/6 1 9/3 1 10/1 1 4/6

Figure 7. Editing of nad3 and rps72 Sequences on Transcript of 0.9 kb from Polysomal Fraction.

Represented here are the sequences of cDNA clones specifically obtained from the 0.9-kb transcript present in the polysomal-enriched RNA fraction. The notations are the same as given in Figures 5 and 6.

Transcripts in the Polysomal-Enriched Fraction Are Mainly Completely Edited

Considering the hypothesis that only completely edited transcripts are utilized for translation, cDNA clones from a wheat mitochondrial polysomal RNA-enriched fraction were obtained and sequenced. The wheat mitochondrial polysomal fraction was obtained as described in Methods after pelleting of the mitochondrial membrane fraction and sedimentation of the supernatant through a 1.5 M sucrose cushion. Different times of centrifugation were tested, and the corresponding fractions were analyzed by RNA gel blot hybridization. RNA extracted from fractions pelleted after 2-hr centrifugation contained nad3-rps72 0.9-kb RNA but much less 3.0-kb precursor RNA and was utilized as a source of polysomal mtRNA. After RNA fractionation on sucrose gradient, nad3-rps 12 cDNA clones were obtained from aliquots containing the polysomal 0.9-kb transcript. From 35 clones that were sequenced, Figures 7 and 8C show that most are completely edited, but partially edited clones were still found. Interestingly, in four of these clones (clones of sequences T6 and T I I) , partia1 editing only affects silent positions.

Deletions of a Nucleotide at Editing Positions

Upon analysis of severa1 cDNA clones, it was found some- times that deletion of a single nucleotide occurred, as shown in Figure 9. It is possible that such deletions have been introduced during PCR amplification, but this is improbable because such deletions have been found

consistently at editing sites and not elsewhere, eight of nine deletions affecting an editing site. These deletions were found once in the nad3 editing site 7 (Figure 9A), five times in site 15 (Figure 9B), and twice in rpsl2 editing site 1 (Figure 9C). The precise nucleotide deleted cannot al- ways be precisely identified, but in each case it was either the edited one or a neighboring T. It is therefore possible to postulate that these deletions are a consequence of the editing process and not an artifact introduced during cDNA synthesis and PCR amplification.

DISCUSSION

The nad3-rps72 Transcription Unit Also Codes for an Unidentified Mitochondrial Protein

The nad3 and rps72 genes are cotranscribed with two additional ORFs. No evidence was found for translation of orf299. The orf756 gene is, however, expressed in wheat mitochondria as a protein associated with the mitochon- drial membrane fraction. Sequences homologous to this gene have been found in many plant mitochondrial ge- nomes analyzed, and it seems, therefore, that it is part of the “core” genetic information of plant mitochondria. In wheat, transcripts for this gene are 10- to 20-fold more abundant than the nad3-rps72 0.9-kb mRNA, possibly as a result of different RNA stabilities or of independent transcription initiation. The presence of a double-loop structure at the 3‘ extremities of the orf756 transcripts could account for their higher stability (Schuster et al.,

11 16 The Plant Cell

A 304 30 i

B O 2 4 6 8 1 0 1 2 1 4 1 6 O 2 4 - 6

301 301 4 4

~ O 2 4 6 8 1 0 1 2 1 4 1 6 O 2 4 6

Number of amino acid conversions Figure 8. Effect of Partia1 Editing on Deduced NAD3 and RPS12 Protein Sequences.

Editing of nad3 and rps72 sequences results in a maximum of 16 and six amino acid conversions, respectively. (A) The histogram represents the number of conversions observed in the sequences of cDNA clones obtained from the 3.0-kb transcript from total mtRNA. (B) The histogram represents the number of conversions observed in the sequences of cDNA clones obtained from the 0.9-kb transcript of total mtRNA. (C) The histogram represents the number of conversions observed in the sequences of cDNA clones obtained from the 0.9-kb transcript from polysomal-enriched mtRNA.

1986). Because no sequence homologies with orf756 can be found in protein data banks, this is an indication that the corresponding putative protein in nonplant mitochon- dria is not encoded by the mitochondrial genomes, some of which have been completely sequenced. The orf756- encoded polypeptide is, therefore, either a mitochondrial protein specific to higher plants or a new example of a nuclear-encoded protein in mammals and fungi that is mitochondrially encoded in higher plants.

Editing 1s Sequence Specific and Translation lndependent

All plant mitochondrial mRNAs analyzed to date, including the nad3, rps72, and orf756 transcripts, are more or less affected by editing. On the contrary, more than 700 tran- scribed nucleotides of the apparently untranslated off299 sequence are not edited. This observation is consistent with the idea that editing in plant mitochondria is specific for translated sequences: rRNA genes and tRNA genes have still not been found to be edited. However, 3 nucleo- tides of the coxll insertion sequence are edited, the same that are also edited in the homologous sequence of the wheat coxll mRNA (Covello and Gray, 1989; Gualberto et al., 1989). This duplicated sequence is, therefore, recog- nized by the factors responsible for editing of the native coxll mRNA, which possibly include complementary tran- scripts (guide RNA, Blum et al., 1990). The fact that the same sequence in different transcripts is identically edited, one being translated and the other not, implies that the specificity of editing is achieved in a strictly sequence- dependent way, independent of transcript translation.

It was previously shown that different editing sites in independent transcripts share sequence similarities (Gualberto et al., 1990b). Sequence similarities have also been found between editing sites of the currently analyzed genes, in addition to the ones already published. Editing sites 1 and 14 of nad3 (codons 2 and 72) share the same YCGGAA motif, editing sites 18, 19, and 21 of nad3 (codons 92, 106, and 1 15) share the same GGAUC motif, and editing sites 3 and 4 of rps72 (codons 74 and 90) contain the same AUUCGCNAGG sequence. Also, editing site 5 of rps72 (codon 95) and editing site 1 (codon 7) of the wheat atp9 gene (Bégu et al., 1990) are preceded by the same AGGUGYNAAAUC sequence. These common sequences could be involved in base pairing with some putative guide RNA.

Editing of nad3 and rpsl2 1s Correlated with RNA Maturation

The study of partially edited transcripts has been of primary importance for the understanding of the trypanosomal

RNA Editing and Maturation in Wheat Mitochondria 1117

T C T C T C T C T C T C

-!- "Ill- c-*."

Figure 9. Deletions of a Nucleotide at Editing Positions.

The figure shows the T and C sequences of three DMA regionswhere deletions of a nucleotide at editing positions were observedin cDNA clones. In each case, the left sequence corresponds toan edited cDNA with edited nucleotides (a T in cDNA sequencecorresponding to a C in the genomic sequence) indicated by dotsand the right sequence corresponds to a clone where at theposition indicated by a triangle the genomic C is missing and thesequence is shorter by 1 base.(A) Partial sequence of cDNA clone T4 (Figure 7) from nucleotide87 to 127 (Figure 2) around editing site 7 of nad3.(B) Partial sequence of cDNA clone M7 (Figure 6) from nucleotide233 to 285 (Figure 2) around editing site 15 of nad3.(C) Partial sequence of cDNA clone M11 (Figure 6) from nucleotide490 to 517 (Figure 2) around editing site 1 of rps12.

former contained significantly more edited nucleotides thanthe latter (Figures 5, 6, and 8). Very few partially editedclones from the 3.0-kb precursor were identical, indicatingthat the use of PCR allows the amplification of independentcDNA clones. The significant differences between the ex-tent of editing of precursor and mature transcripts impliesthat RNA editing in plant mitochondria is a process with atemporal development correlating with transcript matura-tion. This result is up to now the best evidence that editingis a post-transcriptional process.

Accumulation of precursor transcripts that are possiblynot utilized for translation seems, therefore, to be the mainreason for the persistence of partially edited transcripts onplant mitochondria. The existence of partially edited tran-scripts will be, therefore, only a measure of the ratiobetween precursor and mature RNA concentration, asdemonstrated in this study by comparing the partial editingof nad3-rps12 transcripts (where a comparable amount ofthe 0.9- and 3-kb transcripts can be found in the mito-chondria) with the fully edited orf156 (where the 0.8-kbRNA is present at a considerably higher concentration thanthe 3-kb precursor RNA) (Figure 3). Partially edited tran-scripts could probably be sequenced in this case but onlyafter the screening of a very large number of cDNA clones.In agreement with this idea, for wheat mitochondrial genesthat are mainly coded on abundant mRNAs (cox///, nad4,orf156, and atp9, Begu et al., 1990; Gualberto et al.,1990b; Lamattina and Grienenberger, 1990), no partiallyedited transcript has been detected.

mitochondrial editing system. In trypanosomes, editinginvolves the 3' to 5' addition and deletion of uridines inthe RNA sequences, resulting in the creation of frameshifts and initiation and termination codons (Simpson andShaw, 1989). Therefore, partially edited transcripts in try-panosome mitochondria cannot be translated.

In plant mitochondria, partial editing has a different sig-nificance because no frame shifts are created by RNAediting and, therefore, unedited transcripts could theoreti-cally be translated. The first report on RNA editing alreadysuggested that partially edited transcripts can accumulatein plant mitochondria (Gualberto et al., 1989), and thisassumption was confirmed for the Oenothera nad3 andrps13 genes (Schuster et al., 1990; Wissinger et al., 1990).

The significance of partial editing on the expression ofthe genes so far analyzed is not clear. It could be that, inthis way, plant mitochondria achieve protein sequencediversity allowing flexibility to cope with environmentalfluctuations, but this seems unlikely, considering that par-tial editing frequently affects highly conserved amino acids.

The simple transcription pattern of the wheatnad3-rps12 transcription unit permits the separate studyof precursor and mature transcripts. The sequence of alarge number of cDNA clones either from the 0.9-kb matureRNA or from the 3.0-kb precursor clearly showed that the

Mature Transcripts in the Polysomal-Enriched FractionCode Mainly Unique NADS and RPS12 ProteinSequences

The 3.0-kb precursor transcript that in the total mtRNAfraction is mainly partially edited is still detected in RNAgel blots of polysomal RNA but in considerably loweramounts. Its persistence in the polysomal fraction couldresult from contaminant total mtRNA. Considering thispossibility, the differences in the extent of editing betweentotal and polysomal RNA had been compared only for the0.9-kb mature transcript. Most polysomal cDNA clonessequenced are completely edited and code unique NAD3and RPS12 protein sequences, but some clones have beenobtained corresponding to partially edited transcripts. Al-though it is likely that those clones correspond to totalmtRNA contaminating the polysomal fractions, it must beconsidered that they are translated. Even so, it is possiblethat one transcript can be utilized for translation of acompletely edited reading frame but not a cotranscribedand partially edited one: only two of 35 clones from thepolysomal-enriched fraction have both partially edited nad3and rps12 sequences (clones T18 and T19), comparedwith six of 37 clones from total mtRNA (clones M10, M17,M18, M19, M20, andM21).

11 18 The Plant Cell

The number of amino acid conversions deduced from cDNA sequences is represented on histograms in Figure 8. The distribution of the sequences for the highly edited nad3 gene suggests an editing process with slow initiation and termination steps: the majority of the precursor-spe- cific clones show no or few amino acid conversions result- ing from editing, whereas most mature-specific clones are completely or almost completely edited. For nad3, very few clones showing intermediate degrees of editing could be found. No statistically significant difference between extent of editing of mature transcripts in total mtRNA and mature transcripts from the polysomal fraction could be found. However, as shown in Figure 8, it is apparent that most cDNAs derived from the 0.9-kb transcript in total RNA correspond to transcripts in final steps of editing, whereas in polysomes, from the few cDNA clones that are partially edited, there are as many belonging to the inter- mediate edited class as to the almost completely edited class. This suggests that such clones are a background intrinsic to the methodology utilized and may correspond to contaminating 0.9- or 3.0-kb transcripts from total RNA. Although these results constitute little evidence, they sug- gest that there is a mechanism responsible for the func- tional selection of completely edited sequences for trans- lation by a mechanism not related with RNA size selection. This selection could be obtained if partially edited RNAs are sequestered in protein or ribonucleoprotein complexes (i.e., the “editosome” that will be responsible for the editing process) and cannot escape these complexes and attach to the ribosome before the completion of editing.

Deletions of a Nucleotide at Editing Positions lmply an Editing Mechanism lnvolving Nucleotide Replacement

The nature of the C-U editing in plant mitochondria is still unknown. A simple deamination could account for the C-tU nucleotide conversion. However, the observed phe- nomenon of nucleotide deletion in some cDNA clones can be better explained if we consider that the C+U conver- sion is obtained by nucleotide replacement, with or without rupture of the phosphoribose chain of RNA. If one consid- ers that editing involves cutting, nucleotide substitution, and religation, the RNA chain could, in some cases, be religated after cytidine removal but before uridine addition resulting in an RNA shorter by 1 nucleotide residue. An- other possibility is that editing is obtained by pyrimidine replacement without cutting of the RNA chain by an activity similar to the ones found in some DNA repair mechanisms (Olsen et al., 1989). In that case, apyrimidinic positions would exist as intermediary steps of the editing process. It remains to be demonstrated, however, that the reverse transcriptase would not stop at such positions.

METHODS

Oligonucleotide Primers

The following oligonucleotides were synthesized on an Applied Biosystems DNA synthesizer:

01: 6A6CTC6A6A6AAC6AA6T666CTT 02 : G G C A T C T T C C A T T C A T T T C G A 03 : T T A C T C G A G C A C A A T G C T T T T T G T G G C 04 : G C C G A A T T C T G A A A T C A C T A C A G G C 04 : G C C G A A T T C T G A A A T C A C T A C A G G C 05 : T T T A A G C T T A T T T G A A A T C C A A A T C G 0 6 : A A G A A T T C T A C T A A T T C C A A G 07 : A A C A A G C T T G C T T C T T C G A A T C T C G

mtDNA and mtRNA Analyses

Wheat mtDNA and mtRNA were prepared as described (Gualberto et al., 1988). A Sal1 plasmid library containing all the wheat mitochondrial genome was kindly supplied by F. Quétier and B. Lejeune (Orsay, Paris). DNA fragments selected for sequencing were subcloned in M13mpl8 and M13mpl9 and sequenced by the dideoxy chain termination method. The published sequence will appear in the EMBL nucleotide sequence data base as acces- sion number X59153. Conditions for RNA gel blot analysis of total mtRNA were as described by Sambrook et al. (1989). 32P-labeled DNA probes were prepared by second-strand synthesis of M13 clones or by random priming. Conditions for transcript mapping by S1 nuclease protection and primer extension were as previ- ously described by Gualberto et al. (1990a). Size fractionation of mtRNA before cDNA synthesis was performed on semilogarithmic gradients of 10% to 32% sucrose, using a Beckman SW41.Ti rotor at 38,000 rpm for 15 hr.

Fractionation of Mitochondrial Proteins and lmmunodetection

Sucrose gradient-purified mitochondria were lysed by dilution with PBS buffer (Sambrook et al., 1989) and by four to five cycles of freezing and thawing. The membrane fraction was sedimented by centrifugation (30,00Og), washed three times, and solubilized in PBS buffer containing 0.1 O/O SDS. The soluble mitochondrial frac- tion was precipitated with 70% acetone. lmmunodetection of proteins fractionated by SDS-PAGE and transferred to nitrocellu- lose membranes was performed as described by Burnette (1981), using alkaline phosphatase-conjugated antibodies (Sigma).

Expression of Fusion Proteins in Escherichia coli

DNA fragments containing parts of the orf756 and orf299 se- quences were cloned in the Hindlll site of the pHC305AHindlll vector (Amann et al., 1983; Schmidt et al., 1986), which was a kind gift from Urzula Niesbach-Klosgen (Max Planck Institute, Martinsried, Germany). The corresponding fusion proteins were expressed in the W311 O E. coli strain and purified as described (Niesbach-Klosgen et al., 1990). After preparative SDS-PAGE purification, the fusion proteins were electroeluted and injected

RNA Editing and Maturation in Wheat Mitochondria 11 19

into rabbits to raise specific antisera. Preimmune serum was used as a control in immunodetection experiments.

cDNA Synthesis and Amplification

mtRNA was treated with RNase-free DNase (Pharmacia) and the cDNA first strand synthesized using specific primers as described previously by Gualberto et al. (1990b). The resulting cDNA was amplified by PCR using Ta9 DNA polymerase from Pharmacia, following the conditions of the supplier. The amplified fragments were cloned in M13 vectors and sequenced.

Purification of Polysomal-Enriched mtRNA

Wheat mitochondria were purified on sucrose gradients in the presence of 10 mM MgClp and lysed by the addition of 3 volumes of TKM buffer (40 mM Tris-HCI, pH 8.5, 20 mM KCI, 10 mM MgCI2) and 1.5% Triton X-1 00. The mitochondrial membranes were sedimented at 30,OOOg for 30 min, and the soluble fraction was layered over a cushion of 1.5 M sucrose on TKM buffer. Polysomes were sedimented by centrifugation at 38,000 rpm (Beckman SW-41.Ti rotor) for 2 hr. The resulting pellet was resuspended, and contaminating membranes were sedimented by centrifugation at 12,0009. Total RNAs from the soluble fraction were purified by phenol-chloroform extraction and precipitation by 2 M LiCI.

ACKNOWLEDGMENTS

We thank Prof. Jacques-Henry Weil for his continua1 interest and support and Andrée Koch for technical assistance. J.M. Gualberto was supported by a fellowship from the French "Ministère des Affaires Etrangères" (Paris, France), and L.L. was supported by a fellowship from CONICET (Buenos Aires, Argentina) and from CNRS (Paris, France).

Received May 22, 1991 ; accepted August 22, 1991.

REFERENCES

Amann, E., Brosius, J., and Ptashne, M. (1983). Vectors bearing a hybrid trp-lac promoter useful for regulated expression of cloned genes in Escherichia coli. Gene 25, 167-1 78.

Bégu, D., Graves, P-V., Domec, C., Arselin, G., Litvak, S., and Araya, A. (1990). RNA editing of wheat mitochondrial ATP synthase subunit 9: Direct protein and cDNA sequencing. Plant Cell2, 1283-1 290.

Benne, R., Van den Burg, J., Brakenhoff, J., Sloof, P., Van Boom, J.H., and Tromp, M.C. (1986). Major transcript of the frameshift coxll from trypanosome mitochondria contains four nucleotides that are not encoded in the DNA. Cell46,819-826.

Bland, M.M., Levings, C.S. 111, and Matzinger, D.F. (1986). The tobacco mitochondrial ATPase subunit 9 gene is closely linked to an open reading frame for a ribosomal subunit. MOI. Gen. Genet. 204, 8-16.

Blum, B., Balakara, N., and Simpson, L. (1990). A model for RNA editing in kinetoplastid mitochondria: Small "guide R N A molecules transcribed from maxicircle DNA provide the edited sequence information. Cell60, 189-1 98.

Bonen, L., Boer, P.H., and Gray, M.W. (1984). The wheat cyto- chrome oxidase subunit II gene has an intron insert and three radical amino acid changes relative to maize. EMBO J. 3,

Burnette, W.H. (1 981). Western blotting: Electrophoretic transfer of proteins from SDS-polyacrylamide gels to unmodified nitro- cellulose and radiographic detection with antibody and radioio- dinated protein A. Anal. Biochem. 112, 195-203.

Covello, P.S., and Gray, M.W. (1989). RNA editing in plant mitochondria. Nature 341, 662-666.

Finnegan, P.M., and Brown, G.G. (1990). Transcriptional and post-transcriptional regulation of RNA levels in maize mitochon- dria. Plant Cell 2, 71-83.

Gualberto, J.M., Wintz, H., Weil, J.H., and Grienenberger, J.M. (1 988). The genes coding for subunit 3 of NADH dehydrogenase and for ribosomal protein S12 are present in the wheat and maize mitochondrial genomes and are co-transcribed. MOI. Gen. Genet. 215,118-127.

Gualberto, J.M., Lamattina, L., Bonnard, G., Weil, J.H., and Grienenberger, J.M. (1 989). RNA editing in wheat mitochondria results in the ConServation of protein sequences. Nature 341,

Gualberto, J.M., Domon, C., Weil, J.H., and Grienenberger, J.M. (1 990a). Structure and transcription of the gene coding for subunit 3 of cytochrome oxidase in wheat mitochondria. Curr. Genet. 17, 41-47.

Gualberto, J.M., Weil, J.H., and Grienenberger, J.M. (1990b). Editing of the wheat coxlll transcript: Evidence for twelve C to U and one U to C conversions and for sequence similarities around editing sites. Nucl. Acids Res. 18, 3771-3776.

Hiesel, R., Schobel, W., Schuster, W., and Brennicke, A. (1 987). The cytochrome oxydase subunit I and 111 genes of Oenofhera mitochondria are transcribed from identical promoter se- quences. EMBO J. 6,29-34.

Hiesel, R. , Wissinger, B., Schuster, W., and Brennicke, A. (1989). RNA editing in plant mitochondria. Science 246,

Lamattina, L., and Grienenberger, J.M. (1 990). RNA editing of the transcript coding for subunit 4 of NADH dehydrogenase in wheat mitochondria. Uneven distribution of the editing sites among the four exons. Nucl. Acids Res. 19,3275-3282.

Lamattina, L., Weil, J.H., and Grienenberger, J.M. (1989). RNA editing at a splicing site of NADH dehydrogenase subunit IV gene transcript in wheat mitochondria. FEBS Lett. 258, 79-83.

Mahendran, R., Spottswood, M.R., and Miller, D.L. (1991). RNA editing by cytidine insertion in mitochondria of Physarum polycephalum. Nature 349, 434-438.

2531 -2536.

660-662.

1632-1 634.

1 120 The Plant Cell

Mulligan, R.M., Leon, P., and Walbot, V. (1991). Transcriptional and posttranscriptional regulation of maize mitochondrial gene expression. MOI. Cell. Biol. 11, 533-543.

Niesbach-Klosgen, U., Guilley, H., Jonard, G., and Richards, K. (1990). lmmunodetection in vivo of beet necrotic yellow vein virus-encoded proteins. Virology 178, 52-61.

Olsen, L.C., Aasland, R., Wittner, C.U., Krokan, H.E., and Helland, D.E. (1 989). Molecular cloning of human uracil-DNA glycosylase, a highly conserved DNA repair enzyme. EMBO J.

Quagliariello, C., Saiardi, A., and Gallerani, R. (1990). The cytochrome oxidase subunit 111 gene in sunflower mitochondria is cotranscribed with an open reading frame conserved in higher plants. Curr. Genet. 18,355-363.

Quétier, F., Lejeune, B., Delorme, S., Falconet, D., and Jubier, M.F. (1985). Molecular form and function of the wheat mito- chondrial genome. In Molecular Form and Function of the Plant Genomes, NATO AS1 Series A, Life Sciences, Vol. 83, L. van Vloten-Doting, G.S.P. Groot, and T.C. Hall, eds (New York: Plenum Press), pp. 413-420.

Sambrook, J., Fritsch, E., and Maniatis, T. (1989). Molecular Cloning: A Laboratory Manual. (Cold Spring Harbor, NY: Cold Spring Harbor Laboratory).

8,3121-3125.

Schmidt, J., John, M., Wieneke, U., Krüssmann, H-D., and Schell, J. (1986). Expression of the nodulin gene nodA in Rbizobium meliloti and localization of the gene product in the cytosol. Proc. Natl. Acad. Sci. USA 83, 9581-9585.

Schuster, W., Hiesel, R., Isaac, P.G., Leaver, C.J., and Brennicke, A. (1 986). Transcript termini of messenger RNAs in higher plant mitochondria. Nucl. Acids Res. 14, 5943-5954.

Schuster, W., Wissinger, B., Unseld, M., and Brennicke, A. (1 990). Transcripts of the NADH-dehydrogenase subunit 3 gene are differentially edited in Oenotbera mitochondria. EMBO J. 9,

Simpson, L., and Shaw, J. (1989). RNA editing and the mito- chondrial cryptogenes of kinetoplastid protozoa. Cell 57,

Wissinger, B., Hiesel, R., Schuster, W., and Brennicke, A. (1 988). The NADH-dehydrogenase subunit 5 gene in Oenotbera mitochondria contains two introns and is co-transcribed with the 5s rRNA gene. MOI. Gen. Genet. 212,56-65.

Wissinger, B., Schuster, W., and Brennicke, A. (1990). Species- specific RNA editing patterns in the mitochondrial rpsl3 tran- scripts of Oenotbera and Daucus. MOI. Gen. Genet. 224,

263-269.

355-366.

389-395.

DOI 10.1105/tpc.3.10.1109 1991;3;1109-1120Plant Cell

J M Gualberto, G Bonnard, L Lamattina and J M Grienenbergerediting and mRNA maturation.

Expression of the wheat mitochondrial nad3-rps12 transcription unit: correlation between

 This information is current as of February 16, 2020

 

Permissions 8X

https://www.copyright.com/ccc/openurl.do?sid=pd_hw1532298X&issn=1532298X&WT.mc_id=pd_hw153229

eTOCs http://www.plantcell.org/cgi/alerts/ctmain

Sign up for eTOCs at:

CiteTrack Alerts http://www.plantcell.org/cgi/alerts/ctmain

Sign up for CiteTrack Alerts at:

Subscription Information http://www.aspb.org/publications/subscriptions.cfm

is available at:Plant Physiology and The Plant CellSubscription Information for

ADVANCING THE SCIENCE OF PLANT BIOLOGY © American Society of Plant Biologists