• chloroplast transcription;
  • phosphorylation;
  • plastid rpo genes;
  • rifampicin;
  • RNA polymerase subunits


  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

We previously identified two multisubunit plastid RNA polymerases termed A and B. The B enzyme has a bacterial-type polypeptide composition and is sensitive to the prokaryotic transcription inhibitor rifampicin (Rif); the A enzyme has a more complex subunit structure and is Rif-resistant. Here we report results of N-terminal sequencing and MS carried out with the A enzyme, which establish that the latter contains rpo gene products and is structurally related to the B enzyme. Furthermore, evidence is provided that the A enzyme can be converted into a Rif-sensitive enzyme form in a phosphorylation-dependent manner in vitro by a treatment that results in depletion of a β-like subunit. Database searches using sequence information derived from additional polypeptides that are present in purified A preparations revealed sequence similarity with chloroplast proteins involved in RNA processing and redox control. This proteomics approach thus points to the complexity of the chloroplast transcription apparatus and its interconnections with post-transcriptional and signalling mechanisms.


electrophoretic mobility shift assay

cp-pol A (PEP-A)

form A multisubunit plastid RNA polymerase

cp-pol B (PEP-B)

form B multisubunit plastid RNA polymerase


nuclear-encoded phage-type plastid RNA polymerase


bacterial-type plastid RNA polymerase with core subunits encoded by organellar genes


protein kinase A (catalytic subunit)


plastid transcription kinase




superoxide dismutase


(PEP-A-associated factors)




matrix-assisted laser desorption/ionization-MS

Many metabolic functions of plants depend on light as an environmental signal. For instance, following exposure of dark-grown (etiolated) seedlings, they start greening and their cotyledons become converted into photosynthetic organs. At the subcellular level, this is reflected by the photoconversion of the nongreen plastids known as etioplasts into fully active chloroplasts, with accompanying changes in ultrastructure, protein composition, and gene expression patterns [1]. Underlying control processes have mostly been assigned to the level of plastid RNA stability and translation [2,3], but accumulating evidence is available also for transcriptional control [4,5].

Recent progress in the analysis of the structure and function of the plastid transcription apparatus has revealed that at least two distinct DNA-dependent RNA polymerases are involved. One of them, which has long been known [6], is a multisubunit enzyme similar to those in bacteria and in the nuclei of eukaryotic cells. The other enzyme, which was detected more recently [7], is a single-subunit polymerase related to those of bacteriophages T3 and T7 and mitochondria [5,8,9]. The multisubunit enzyme was termed PEP (plastid-encoded polymerase) because of the intraorganellar coding sites of its core subunits, whereas the phage-type enzyme is nuclear-encoded [10–14] and hence was named NEP [5].

In mustard (Sinapis alba L.), we identified two distinct multisubunit plastid RNA polymerases, named enzymes A (cp-pol A) and B (cp-pol B), which differ in terms of subunit composition, functional properties, and abundance during etioplast[RIGHTWARDS ARROW]chloroplast conversion [15]. The type B RNA polymerase consists of four polypeptides (154, 120, 78 and 38 kDa) that match the predicted sizes of the plastid rpoC2, rpoB, rpoC1 and rpoA gene products from other plant species (reviewed in [16–18]). The type A enzyme is larger than the B enzyme, consisting of at least 13 putative subunits. Whereas the B enzyme is sensitive to the prokaryotic transcription inhibitor rifampicin (Rif) [19–21], the A enzyme is resistant to the drug. The latter enzyme form predominates in chloroplasts, whereas the B form is the major activity in etioplasts and in immature plastids during greening [15,22]. It was suggested, therefore, that the two enzyme forms may be structurally related, with a possibility of B[RIGHTWARDS ARROW]A interconversion during chloroplast development [23].

To help clarify the possible structural and functional relationships of the A and B enzymes, two questions were addressed in the present work: (a) whether or not these two polymerases share common features at the level of the primary structure of their polypeptides; and (b) what might be the basis for their distinct properties, including Rif sensitivity vs. resistance.

Materials and methods

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

DNA sequencing

Chloroplast DNA regions containing rpo sequences were cloned into pBS (Stratagene). Dideoxy DNA sequencing [24] was carried out both manually and using an ALF automatic sequencer (Amersham Pharmacia). Sequence data have been submitted to the EMBL databases under the accession numbers X82417, Y13690, AJ243345, and AJ243754.

Purification of plastid RNA polymerase; activity assay

The standard purification scheme for the A and B enzyme preparations from 5-day-old mustard seedlings included chromatography of plastid lysates on heparin–Sepharose CL6B, followed by glycerol gradient density centrifugation, essentially as described [15,22]. In vitro transcription activity was assayed by trichloroacetic precipitation and scintillation counting of radiolabelled transcripts as described [25].

Two-dimensional gel electrophoresis of the chloroplast polymerase A

First-dimension separation was in a native electrophoretic mobility shift assay (EMSA) gel [26] after incubation of purified form A enzyme with a psbA promotor fragment prepared from pSA05/H120 [27]. The fragment was 3′-labelled by fill-in with α-32P-dATP (Amersham) and Klenow enzyme (BRL), and 2.5 ng of the labelled probe were used as described [22]. The sample (100 µL total volume) was loaded on a 4.5% native polyacrylamide tube gel (2.5 mm diameter, 150 mm length) and electrophoresed at 0.5 mA for 5 h. The gel was exposed to X-ray film at 4 °C for 2–4 h and was then frozen at −20 °C overnight in a buffer containing 4% SDS, 5% 2-mercaptoethanol and 60 mm Tris/HCl pH 6.8. After thawing, the tube gel was incubated in the buffer at 20 °C for 30 min, washed twice with 0.4% SDS and 60 mm Tris/HCl pH 6.8 to remove excess 2-mercaptoethanol, and layered horizontally on to a 5–15% SDS polyacrylamide gel (300 × 170 × 1.0 mm) [28]. After electrophoresis (15 h at 14 mA), proteins were stained with silver [29]. The gel was dried and autoradiographed for detection of the radioactive psbA fragment.

Preparation of polymerase A for protein sequencing and MS

The purification scheme using cotyledons from 4-day-old light-grown mustard seedlings involved chromatography on heparin–Sepharose (Pharmacia) followed by glycerol gradient centrifugation as described [15,22]. Polymerase A from the pooled gradient fractions at 23–27% glycerol (3.5 mL; 20 µg protein) was concentrated by precipitation in 80% methanol and then analysed on a denaturing 5–15% SDS polyacrylamide gel [28]. Separated polypeptides were transferred to a poly(vinylidene difluoride) membrane (Immobilon PSQ; Millipore) and stained with Coomassie brilliant blue R-250. Protein bands were excised and subjected to automated Edman degradation in a protein sequencer (Applied Biosystems Model 476A) according to Meyer et al.[30].

For MS, 2.5 mL glycerol gradient-fractionated A enzyme (15 µg) was methanol-precipitated and subjected to SDS/PAGE as described above. After electrophoresis polypeptides were stained directly without blotting, excised, and subjected to in-gel digestion with trypsin. Gel pieces were washed in 10 mm NH4HCO3 (pH 7.8), followed by acetonitril/10 mm NH4HCO3 (1 : 1 v/v, pH 7.8), three times for 10 min each. They were shrunken in acetonitrol and rehydrated in 2 µL trypsin solution consisting of 0.05 mg·mL−1 sequencing grade modified trypsin (Promega V5111) in 10 mm NH4HCO3 (pH 7.8). After digestion at 37 °C for 8 h, the supernatant was collected. Each gel piece was then reextracted with 5 µL 10 mm NH4HCO3 at 37 °C for 30 min in a sonication bath, and the supernatants were combined and concentrated using ZipTip (Millipore) pipette tips. The eluted peptide fragments were subjected to matrix-assisted laser desorption/ionization-MS (MALDI-MS) according to [31]. The MALDI matrix was a saturated solution of 4-hydroxy-α-cinnamic acid in 0.1% trifluoroacetic acid/acetonitril (1 : 1, v/v). Samples were mixed on target with the same matrix solution. MALDI mass spectra were recorded on a Bruker Reflex III (Bruker Daltonik).

Conversion of chloroplast RNA polymerase A into a Rif-sensitive form

The procedure followed that for RNA polymerase A purification described above, except that the partially purified enzyme was phosphorylated and then run over a phosphocellulose column, before it was subjected to the final glycerol gradient step of the standard procedure. In brief, 2.5 mL (1.2 mg protein) of the heparin–Sepharose-stage material (transcriptionally active fraction; HS-TA) was incubated in the presence of 40 µm unlabelled ATP, 50 µg·mL−1 phenylmethylsulfonylfluoride and 1 mm benzamidine at 30 °C for 30 min as described [32]. Under these conditions, in vitro phosphorylation was catalysed by the plastid transcription kinase (PTK), which copurifies with the RNA polymerase during heparin–Sepharose chromatography [32,33]. The sample was then adjusted to 50 mm ammonium sulfate in 50 mm Tris/HCl, pH 7.6, 0.1 mm EDTA, 5 mm 2-mercaptoethanol, 0.1% (v/v) Triton X-100, 10% (v/v) glycerol. It was loaded onto a 5-mL phosphocellulose column (Whatman P11), washed with the same buffer, and bound protein was then eluted using a linear 0.05–1.6 m ammonium sulfate gradient. The active fractions eluting in a narrow peak at 0.1–0.3 m salt were pooled and subjected directly to glycerol gradient centrifugation essentially as described [15]. Controls included ‘mock’ samples incubated in the absence of ATP as well as those that were dephosphorylated by calf intestinal alkaline phosphatase (CIAP) at the heparin–Sepharose stage [32]. Stock solutions containing 10 mg·mL−1 Rif in methanol were diluted with water immediately before use. Control reactions without Rif contained the same amounts of diluted methanol.

Phosphorylation state assay

Radioactive labelling of RNA polymerase subunits in the presence of γ[32P]ATP and the catalytic subunit of bovine heart protein kinase A (PKA; Sigma) was carried out with 200 µL of glycerol gradient-purified RNA polymerase (20 µg protein) in a final volume of 400 µL [33]. Fifty microlitre portions of the reaction mixture were used for polypeptide analysis by 5–15% SDS/PAGE [28], followed by silver staining [29]. Radioactively labelled polypeptides were detected by autoradiography of the dried gels.


  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Plastid rpo genes and polymerase subunits from mustard

Both the rpoA gene and the rpoBC1C2 cluster on mustard chloroplast DNA were previously cloned and sequenced in our group (EMBL accession number AJ243345 for rpoA and AJ243754 for rpoBC1C2). The derived sizes of the rpo gene products, i.e. 37.8 kDa (α), 121.6 kDa (β), 80 kDa (β′) and 157.3 kDa (β″), are in agreement with the apparent molecular masses of the subunits of purified plastid RNA polymerase B from mustard (Fig. 1, panel B). Polymerase A preparations (Fig. 1, panel A), however, contain a larger number of polypeptides and, with the possible exception of a band at ≈ 78 kDa, the migration positions of the A polypeptides do not match precisely those of the B subunits. We therefore initiated experiments to gain information on the identity of the A subunits.


Figure 1. Plastid rpo genes and multisubunit RNA polymerases A and B from mustard. The sequenced rpo genes (open bars) were drawn to scale (size bar in lower right corner). Features shown include the Rif clusters of rpoB (roman numerals; see Fig. 5 for details) as well as the rpoC1 intron (hatched). (A) and (B). Silver-stained polypeptides of plastid RNA polymerases A and B, respectively, separated by SDS/PAGE on a 5–15% gel. Each rpo gene is assigned to a corresponding subunit (α, β, β′ and β″) of the B enzyme. Marker proteins (kDa values at left margin) were, from top to bottom: α2-macroglobulin (170 kDa), β-galactosidase (116 kDa), transferrin (76 kDa), glutamic dehydrogenase (53 kDa), aldolase (39 kDa), chymotrypsin (25 kDa) and cytochrome C (12.5 kDa).

Download figure to PowerPoint

To further resolve the A polypeptides, we used a two-dimensional gel electrophoretic approach (Fig. 2). First-dimension separation was carried out using native PAGE as part of an EMSA experiment [26], i.e. following incubation of the A enzyme with a DNA fragment that carries the chloroplast psbA promoter [27]. The native gel was then used for second-dimension polypeptide analysis by SDS/PAGE.


Figure 2. Two-dimensional gel separation of chloropast RNA polymerase A. Upper panel (1. D) shows an autoradiograph of the first-dimension native EMSA gel, following incubation of purified enzyme A with a radioactively labelled DNA fragment that contains the mustard chloroplast psbA promoter. Positions of the free DNA (f) and the DNA–protein complex (a) are marked. Lower panel (2. D) shows silver-stained polypeptides after second-dimension SDS/PAGE in a 5–15% gel. The kDa values of the polymerase A-associated polypeptides (‘A’) are given in the left margin and those of marker proteins (see Fig. 1) are shown on the right.

Download figure to PowerPoint

As is evident from Fig. 2, at least 13 bands that had previously been assigned as components of the purified A enzyme [15] are detected in the DNA–protein complex. Two polypeptides at 76 and 75 kDa, which were unresolved in one-dimensional SDS/PAGE (Fig. 1), are clearly visible. Control two-dimensional gel electrophoresis of the A enzyme preparation alone (in the absence of DNA) revealed an identical stained pattern (data not shown), suggesting that all polypeptides present in the purified fraction were true constituents of polymerase A.

Identification of polymerase A polypeptides by microsequencing and MS

Considering differences in number and size of the polymerase A and B polypeptides, it was of interest to find out whether or not the two polymerase forms were related at the level of polypeptide primary sequence and what might be the extra polypeptides of the A enzyme. To address these questions, we set out to determine the N-terminal sequences and mass fingerprints.

Following SDS/PAGE and blotting to poly(vinylidene difluoride) membranes, we obtained N-terminal amino-acid sequences for the 141, 110, 36 and 26-kDa polypeptides and identified them by database homology searches (http://www. Other polypeptides either did not yield sequence information at all, probably because they were N-terminally blocked, or the obtained sequences were too short and/or did not give unambiguous results (those at 107, 78 and 53 kDa). As shown in Fig. 3, the N-terminal peptides determined by microsequencing of the 141-kDa and 110-kDa subunits match regions of the derived amino acid sequences of the mustard rpoC2 and rpoB genes, respectively. This suggests that these two RNA polymerase A subunits are indeed structurally related to the β″ and β core constituents of the B enzyme (Fig. 1).


Figure 3. N-terminal sequence alignments ofRNA polymerase A polypeptides. Amino acid sequences obtained for the 141, 110, 36 and 26-kDa polypeptides (Figs 1 and 2) are shown in bold letters in lane 1 of each alignment. 141 kDa, lanes 2–7: derived amino acid sequences of the rpoC2 gene for the β″ subunit of the plastid RNA polymerase from Sinapis alba (S.a.; accession number X82417), Arabidopsis thaliana (A.t.; accession number Y13690), tobacco (N.t.) [56]), spinach (S.o.) [57], maize (Z.m.) [58], and rice (O.s.) [59]. 110 kDa, lanes 2–7: derived sequences of the rpoB gene for the β subunit from S. alba (accession number X82417), A. thaliana (Y13690) and the four other higher plants. Underlined residues from maize were confirmed by protein sequencing [37]. Numbers to the left and right indicate amino acid positions within each derived sequence. Dots: identical residues; ‘–’: positions missing in one or more sequences. 36 kDa, lanes 2–4: derived N-terminal regions of ESTs from A. thaliana (A.t., accession numbers Y10557 and Y15382) and rice (O.s, accession number C23596) and of the SynechococcusPCC6803 gene slr1540 (accession number D90906). These sequences reveal significant overall similarity with CSP41, an RNA binding protein from spinach chloroplasts (accession number U49442[36]; not shown). Boxed region: putative cleavage site [34]. 26 kDa, lanes 2–4: regions of derived sequences of Fe-SOD genes from Glycine max (G.m. [60]), Zantedeschia aethiopica (Z.a., accession number AF094831), and A. thaliana (A.t., AF061852). Boxed: putative cleavage site.

Download figure to PowerPoint

The residues read from the N-terminus of the 36-kDa band can be aligned with a region of the derived amino acid sequence of two expressed sequence tag (EST) sequences from Arabidopsis and rice, located immediately downstream of a putative cleavage site (VXA) [34]. Both EST-derived sequences reveal N-terminal extensions that exhibit typical features of a chloroplast transit peptide, i.e. clustered serines and a high number of positively charged amino acids [35]. In addition, the N-terminal residues of the mustard 36-kDa polypeptide resemble those derived from a cyanobacterial gene (Synecchococcus PCC6803; slr1540), which lacks the putative transit peptide region present in the sequences from Arabidopsis and rice. blast homology searches using the derived amino acid sequences of the two higher plant ESTs and the cyanobacterial gene revealed 34–38% identical and 54–58% conservative amino acid positions with CSP41, a nuclear-endcoded RNA binding protein from spinach chloroplasts [36]. The score (160) was more than twice than that of the next best (70) of all other matches, suggesting that there is a significant relationship between these sequences (data not shown).

The N-terminus of the 26-kDa band (Fig. 3) is highly similar to that derived from the nucleotide sequences of mature iron superoxide dismutase (Fe-SOD) from chloroplasts of other higher plant species. As in the case of the 36-kDa polypeptide, this region is located just behind a putative cleavage site, which is preceded by N-terminal extensions with transit peptide signatures for each of the three SODs.

In a second approach towards identification of polymerase A polypeptides we determined mass fingerprints by MALDI-MS, followed by database analyses using ProFound ( The results obtained are summarized in Table 1. In brief: (a) these data confirmed that the 110-kDa polypeptide is likely to represent the rpoB-encoded catalytic β subunit; (b) unexpectedly, the same degree of similarity was found also for the 107-kDa band, suggesting that this polypeptide may represent a modified β form; (c) The mass fingerprint of the 36-kDa band confirmed the results from Edman degradation (Fig. 3), i.e. it matches the derived protein sequences of Arabidopsis EST sequences that are related to the RNA-binding protein CSP41 [36]; (d) The 29-kDa component was tentatively identified as an annexin-like protein.

Table 1. Tentative assignment of polymerase A polypeptides by MALDI-MS. RNA polymerase was purified by heparin–Sepharose chromatography and glycerol gradient density centrifugation, polypeptides were then separated by SDS/PAGE, followed by MS and database searches as described. The values given for sequence coverage take into account possible methionine oxidation, deamidation of glutamine, and acrylamide modification of free cysteine residues. A mass tolerance of 0.15 Da was accepted. All listed proteins were from A. thaliana.
Size (kDa)Sequence coverage (%) Tentative protein Accession number
11047DNA-dependent RNA polymerase β subunitY13690
10746DNA-dependent RNA polymerase β subunitY13690
 3648Putative RNA-binding proteinY15382, Y10557
 2951Annexin-like proteinAC005499

Taken together, the results of subunit identification by microsequencing and MALDI-MS (Fig. 3 and Table 1) indicate that the form A enzyme indeed contains polypeptides with characteristics of rpo gene products, i.e. the β″-like band at 141 kDa as well as the two β-like bands at 110 and 107 kDa. In addition, three polymerase A-associated polypeptides were tentatively identified. The latter showed similarity with nuclear-encoded proteins, each of which can be assigned to a previously unexpected but physiologically important function in chloroplast transcription (see Discussion).

Conversion of RNA polymerase A into a Rif-sensitive enzyme form

In eubacterial systems, the catalytic β subunit is known to be the target for Rif binding [19–21]. The presence of two different β-like polypeptides in chloroplast RNA polymerase A could be related to our previous observations that the A enzyme is Rif-resistant, whereas the B enzyme is sensitive to the drug [15]. We therefore tested if it was possible to reconvert the A enzyme into a Rif-sensitive form in vitro. A procedure that proved to be effective is outlined in Fig. 4A. Partially purified RNA polymerase (heparin–Sepharose stage) was phosphorylated by taking advantage of the PTK activity that copurifies with the polymerase At this stage [32,33]. The kinase-treated polymerase was then subjected to phosphocellulose chromatography, followed by the final glycerol gradient centrifugation step of the standard purification scheme. As is evident from Fig. 4B, the phosphorylation and phosphocellulose treatment resulted in an enzyme preparation (lane 4) that was more than 50% inhibited at a Rif concentration of 1 µg·mL. None of the controls, including untreated (lane 1), ‘mock’-treated (lane 2) or dephosphorylated enzyme (lane 3), revealed significant Rif inhibition.


Figure 4. Conversion of RNA polymerase A into a Rif-sensitive enzyme form. (A) Purification scheme without (left) or with the conversion treatment, consisting of phosphorylation at the heparin–Sepharose stage, followed by phosphocellulose chromatography (right). (B) Transcription activity of chloroplast RNA polymerase in the presence of Rif. Preparations 1–4 all were obtained by the procedure including the phosphocellulose step (A). They differ, however, with regard to the treatment used at the heparin–Sepharose stage: (1) no treatment (100% control) (2) incubation under nonphosphorylation conditions in the absence of ATP (‘Mock’) or (3) in the presence of CIAP (4) phosphorylation by the endogenous PTK present in the heparin–Sepharose fraction [33]. Values shown were in the presence of Rif (1 µg·mL−1) relative to those in the absence of the drug. The 100% value corresponds to 2885 c.p.m. of trichloroacetic acid-precipitable material in the transcription assays. To account for differences in specific activity (up to 20%) of the preparations 1–4, their amounts were adjusted to result in equal activity in the absence of Rif. Values represent means of three or four independent experiments, each of which was performed in duplicate.

Download figure to PowerPoint

We next asked how the phosphorylation and phosphocellulose treatment might have caused the partial interconversion of the A enzyme into a Rif-sensitive form. We analysed the polypeptide composition of the purified enzyme preparations after the final glycerol gradient step (Fig. 4A). Furthermore, to assess the phosphorylation state of individual subunits, the same purified preparations were radioactively labelled using γ[32P]ATP and PKA, followed by SDS/PAGE.

As shown in Fig. 5 (left panel), the CIAP-pretreated A enzyme (lane 2) revealed essentially the same stained pattern as the untreated control (lane 1). In contrast, the preparation that had been pretreated with PTK (lane 3) and was found to be partially Rif-sensitive (Fig. 4B) showed reduced amounts of the 107-kDa β-like polypeptide. This effect was noticeable, however, only if the intermittent phosphocellulose step (Fig. 4A) was included in the purification scheme. Omission of this step resulted in a Rif-resistant A enzyme that contained normal amounts of the 107-kDa band as compared to the preparations depicted in Figs 1 and 2 (data not shown).


Figure 5. Polypeptide and phosphorylation analysis of modified RNA polymerase. HS-stage enzyme (Fig. 4A) remained untreated (‘control’; lanes 1 and 5) or was incubated either under dephosphorylation (‘CIAP’, lanes 2 and 6) or phosphorylation conditions (′PTK’, lanes 3 and 7) as in Fig. 4B, followed by the phosphocellulose and glycerol gradient steps. The purified samples were incubated with γ-32P-ATP and PKA, and then separated on a 5–15% SDS/PAGE gel. After silver-staining (left panel, lanes 1–4), gels were dried and autoradiographed (right panel, lanes 5–8). Lanes 4 and 8: PKA alone in the absence of plastid RNA polymerase. Asterisks (from top to bottom): bands at 107 kDa (arrow), 76–72 kDa, PKA (arrow), 35 kDa and 29 kDa mentioned in the text. The kDa values of marker polypeptides (Fig. 1) are given in the right margin.

Download figure to PowerPoint

Figure 5 (right panel), lanes 5–8, shows the autoradiograph corresponding to the stained patterns in lanes 1–4. Of the putative subunits that comprise the form A enzyme (Figs 1 and 2), those at 107 kDa, 72–76 kDa (unresolved), 35 kDa and 29 kDa gave the strongest radioactive signals. As in the stained patterns (Fig. 5, lanes 1 and 2), there was no significant difference in relative intensity of labelled bands from the CIAP-pretreated A enzyme (lane 6) compared to the untreated control (lane 5), whereas the 107-kDa signal was reduced in the sample (lane 7) that had been subjected to the phosphorylation and phosphocellulose treatment (Fig. 4). This pretreatment led to reduced signal intensity also at the positions of the 72–76, 35 and 29-kDa bands (Fig. 5, lane 7). As the stained polypeptides corresponding to these radioactive bands were present in normal amounts (Fig. 5, lane 3) in contrast with the 107-kDa polypeptide, it is likely that their PKA substrate site(s) may have become substantially saturated during the preceding PTK/phosphocellulose treatment. In conclusion, these data indicate that the 107-kDa β-like subunit was preferentially removed from the form A complex by the pretreatment, and this selective loss was correlated with the appearance of Rif sensitivity.


  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

In this study we have obtained evidence, by N-terminal sequencing and MALDI-MS, for the existence of rpo-gene products in the multisubunit chloroplast RNA polymerase A from mustard. The N-terminal sequences of the two largest (141 and 110 kDa) polypeptides were found to match those of the derived rpoC2 and rpoB gene products (Fig. 3), which supports earlier suggestions that the chloroplast polymerase A is structurally related to the bacterial-type B enzyme that predominates in etioplasts [15,22]. In accordance with recent nomenclature [5], we propose to name these two enzyme forms PEP-A and PEP-B, respectively.

Microsequencing revealed that the N-terminal region of the 141-kDa PEP-A subunit from mustard is almost identical to that reported for the largest subunit of purified chloroplast RNA polymerase from maize, i.e. the only chloroplast RNA polymerase for which direct protein sequence information is as yet available [37,38]. On the other hand, the alignment of the derived protein sequences (Fig. 3) showed a short hypothetical extension, MEVL, for the dicot plant species mustard, Arabidopsis, spinach and tobacco, which is absent in the monocot species maize and rice. The perhaps most reasonable explanation would be that the second methionine residue in the dicot species marks the translation start site. However, it cannot be excluded that the extra peptide is synthesized and subsequently removed in vivo. It would be interesting to know if this region might be retained in RNA polymerase B from mustard. Despite various attempts, however, we consistently found the B subunits to be N-terminally blocked (data not shown).

The protein and DNA sequence data for the second largest (110 kDa) PEP-A subunit from mustard (Fig. 3) provide a picture that differs from that for the 141-kDa subunit. The microsequenced peptide aligns further upstream than that of the corresponding maize subunit [37], suggesting the existence of a 6 amino-acid extension in mustard. This is probably the result of the additional methionine residue in front of that used as the start methionine in most of the other plant sequences, except in the closely related crucifer Arabidopsis (Fig. 3).

The mass fingerprints of both the 110-kDa and 107-kDa polypeptides identified both as rpoB-related gene products. The presence of two forms of β subunit in PEP-A may seem peculiar, as in bacterial RNA polymerases β is the essential core subunit that catalyses polymerization of NTPs into RNA. An explanation suggesting a regulatory role comes from the rifampicin experiments (Fig. 4) discussed below. In the current work, we have not yet addressed experimentally the question of what might be the reason(s) for the different migration behaviour of these two sequence-related polypeptides, which could include proteolytic cleavage as well as a range of protein modifications. We note, however, that the 107-kDa but not the 110-kDa band was a substrate for PKA (Fig. 5), whereas neither of them was efficiently phosphorylated by PTK [32], suggesting that the different migration behaviour does not seem to simply reflect differences in phosphorylation state. While detailed analyses are part of our ongoing work, the present data obtained by microsequencing and MALDI-MS help to clarify another pertinent point. They exclude the possibility that either of these two bands might be closely related to the single-subunit NEP enzyme [13,14,39], which was reported to migrate in the 107–110-kDa region in spinach [7].

In addition to rpo gene products, microsequencing and MS identified three PEP-A components that are likely to represent nuclear gene products of diverse functions. The latter polypeptides, at 36 kDa (putative RNA binding protein), 29 kDa (annexin-like protein), and 26 kDa (Fe-SOD-like protein), all were consistently found in highly purified PEP-A preparations, including those that were subjected to two-dimensional gel electrophoretic separation (Fig. 2) or additional column chromatograpy steps. Furthermore, none of the determined sequences revealed contamination by major chloroplast proteins such as rubisco or components of the photosynthetic apparatus (data not shown), suggesting that the bands at 36, 29 and 26 kDa might be true constituents of chloroplast PEP-A.

What might be the functional role of these polymerase-associated components? The presence of a RNA binding protein (putative function of the 36-kDa component; Fig. 3) could be important in the stabilization and/or maturation of nascent transcripts. Both transcription and post-transcriptional RNA processing events are known to be driven by thylakoid-associated enzymatic machineries (reviewed in [1,40,41]). Moreover, evidence is available for components that act at the interface between transcription and RNA processing in a number of systems ranging from bacteria to vertebrates [42–44].

The existence of a Fe-SOD-related polypeptide (26-kDa band) in mustard PEP-A could be viewed as a consequence of thylakoid-associated chloroplast transcription. As photosynthetic electron transport can be a source of oxygen radicals [45,46], there is a need for efficient protection and detoxification mechanisms, in which SODs are known to play central roles [47,48]. One implication of the physical proximity therefore would be that newly synthesized transcripts must be protected from damage by radicals generated as photosynthetic by-products.

The annexin-like 29-kDa protein may have a related function. Plant annexins are calcium-dependent phospholipid-binding proteins that are often membrane-associated [49], which would be consistent with an anchoring role for this polymerase subunit. Furthermore, an Arabidopsis annexin that revealed significant similarity with the 29-kDa polypeptide was previously shown to counteract H2O2 stress in an oxyR-deficient Escherichia coli strain after transformation [50]. This may point to a possible involvement of the 29-kDa protein in radical detoxification, i.e. as scavenger of H2O2 molecules generated by SODs, including the putative Fe-SOD (26-kDa band) of PEP-A (Fig. 3). Furthermore, H2O2 itself (or other reactive oxygen intermediates) might act as a transcriptional signal in chloroplasts, as was previously shown to be the case for oxyR-dependent transcription in bacterial systems [51,52]. We note that chloroplast transcription is modulated by redox-reactive reagents both in vitro[32,53] and in organello (T. Pfannschmidt and G. Link, unpublished data).

As shown in Figs 4 and 5, phosphorylation of PEP-A followed by phosphocellulose chromatography resulted in an enzyme form that had regained significant Rif sensitivity. The stained SDS/PAGE patterns (Fig. 5) revealed depletion of the 107-kDa (β107) subunit, indicating that this polypeptide may be a key determinant for Rif resistance vs. sensitivity. It is conceivable that the presence of this component affects the conformation of the PEP-A core in a way which prevents interference of Rif with transcription initiation, although the detailed mechanisms remain to be established. It seems reasonable to conclude, however, that the phosphorylation/phosphocellulose treatment used (Fig. 4) does not result in an altered phosphorylation state of the β107 component itself. It is not an efficient substrate for PTK, i.e. the endogenous serine-type protein kinase used for the pretreatment of PEP-A [32]. The loss of the β107 subunit thus seems to be an indirect effect in reponse to phosphorylation of other PEP-A polypeptide(s), the most likely candidates of which are the known PTK substrate bands at 72–76 kDa [32].

In terms of our model proposed for the B[RIGHTWARDS ARROW]A conversion during light-induced chloroplast formation [15,23], the ‘nonrpo’ polypeptides could be viewed as AAFs (PEP-A-associated factors) that are recruited into the large A enzyme complex. These AAFs provide new functions to the transcriptional machinery, which may help to adapt and regulate transcription under conditions of increasing photosynthetic activity. Our data support the current view that the chloroplast transcription apparatus is much more complex than could be envisaged from the bacterial-type organization of the rpo genes, which fully account for the α(2)ββ′β″ architecture of the etioplast B enzyme.

Both the number and composition of accessory polypeptides of the transcription complex seem to vary depending on age, tissue type and environmental cues. For instance, different forms of chloroplast RNA polymerase complexes were purified from pea, each of which contained both common and distinct polypeptides and revealed different transcriptional properties [54]. Likewise, the PEP enzyme from wheat was recently reported to show tissue type-specific transcription specificity [55]. Control mechanisms that regulate the recruitment and activity of the polymerase-associated factors are beginning to be elucidated. Further characterization of these factors, preferably in a cloned recombinant form, can be anticipated to help clarify their role during plastid development and function.


  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

The authors thank U. Berchner-Pfannschmidt, K. Jülicher, M. Kestermann and K. Liere for their participation in the rpo gene sequencing work, C. Wittig for expert technical assistance and H. Summer for help with the database searches. This work was funded by the Deutsche Forschungsgemeinschaft (Li 261/14–3 and Me 753/5–3) and the Fonds der Chemischen Industrie, Germany.


  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  • 1
    Herrmann, R.G., Westhoff, P., Link, G. (1992) Chloroplast biogenesis in higher plants. In Cell Organelles (Herrmann, R.G., ed.), pp. 275 349. Springer-Verlag, Wien, Heidelberg, New York.
  • 2
    Rochaix, J.D. (1996) Post-transcriptional regulation of chloroplast gene expression in Chlamydomonas reinhardtii. Plant Mol. Biol. 32, 327 341.
  • 3
    Stern, D.B., Higgs, D.C., Yang, J. (1997) Transcription and translation in chloroplasts. Trends Plant Sci. 2, 308 315.DOI: 10.1016/s1360-1385(97)01088-1
  • 4
    Mullet, J.E. (1993) Dynamic regulation of chloroplast transcription. Plant Physiol. 103, 309 313.
  • 5
    Maliga, P. (1998) Two plastid RNA polymerases of higher plants: an evolving story. Trends Plant Sci. 3, 4 6.DOI: 10.1016/s1360-1385(97)87981-2
  • 6
    Smith, H.J. & Bogorad, L. (1974) The polypeptide subunit structure of the DNA-dependent RNA polymerase of Zea mays chloroplasts. Proc. Natl Acad. Sci. USA 71, 4839 4842.
  • 7
    Lerbs-Mache, S. (1993) The 110-kDa polypeptide of spinach plastid DNA-dependent RNA polymerase: Single-subunit enzyme or catalytic core of multimeric enzyme complexes? Proc. Natl Acad. Sci. USA 90, 5509 5513.
  • 8
    Gray, M.W. & Lang, B.F. (1998) Transcription in chloroplasts and mitochondria: a tale of two polymerases. Trends Microbiol. 6, 1 3.DOI: 10.1016/s0966-842x(97)01182-7
  • 9
    Weihe, A. & Börner, T. (1999) Transcription and architecture of promoters in chloroplasts. Trends Plant Sci. 4, 169 170.DOI: 10.1016/s1360-1385(99)01407-7
  • 10
    Morden, C.W., Wolfe, K.H., DePamphilis, C.W., Palmer, J.D. (1991) Plastid translation and transcription genes in a non-photosynthetic plant: intact, missing and pseudo genes. EMBO J. 10, 3281 3288.
  • 11
    Hess, W.R., Prombona, A., Fieder, B., Subramanian, A.R., Börner, T. (1993) Chloroplast rps15 and the rpoB/C1/C2 gene cluster are strongly transcribed in ribosome-deficient plastids: evidence for a functioning non-chloroplast-encoded RNA polymerase. EMBO J. 12, 563 571.
  • 12
    Allison, L.A., Simon, L.D., Maliga, P. (1996) Deletion of rpoB reveals a second distinct transcription system in plastids of higher plants. EMBO J. 15, 2802 2809.
  • 13
    Hedtke, B., Börner, T., Weihe, A. (1997) Mitochondrial and chloroplast phage-type RNA polymerases in Arabidopsis. Science 277, 809 811.
  • 14
    Chang, C.C., Sheen, J., Bligny, M., Niwa, Y., Lerbs-Mache, S., Stern, D.B. (1999) Functional analysis of two maize cDNAs encoding T7-like RNA polymerases. Plant Cell 11, 911 926.
  • 15
    Pfannschmidt, T. & Link, G. (1994) Separation of two classes of plastid DNA-dependent RNA polymerases that are differentially expressed in mustard (Sinapis alba L.) seedlings. Plant Mol. Biol. 25, 69 81.
  • 16
    Igloi, G.L. & Kössel, H. (1992) The transcriptional apparatus of chloroplasts. Crit. Rev. Plant Sci. 10, 525 558.
  • 17
    Sugiura, M. (1992) The chloroplast genome. Plant Mol. Biol. 19, 149 168.
  • 18
    Sugita, M. & Sugiura, M. (1996) Regulation of gene expression in chloroplasts of higher plants. Plant Mol. Biol. 32, 315 326.
  • 19
    Hartmann, G., Honikel, K.O., Knüsel, F., Nuesch, J. (1967) The specific inhibition of the DNA-directed RNA synthesis by rifamycin. Biochim. Biophys. Acta 145, 843 849.
  • 20
    Jin, D.J. & Gross, C.A. (1988) Mapping and sequencing of mutations in the Escherichia coli rpoB gene that lead to rifampicin resistance. J. Mol. Biol. 202, 45 58.
  • 21
    Severinov, K., Soushko, M., Goldfarb, A., Nikiforov, V. (1993) Rifampicin region revisited. New rifampicin-resistant and streptolydigin-resistant mutants in the β subunit of Escherichia coli RNA polymerase. J. Biol. Chem. 268, 14820 14825.
  • 22
    Pfannschmidt, T. & Link, G. (1997) The A and B forms of plastid DNA-dependent RNA polymerase from mustard (Sinapis alba L.) transcribe the same genes in a different developmental context. Mol. Gen. Genet. 257, 35 44.
  • 23
    Link, G. (1996) Green life: control of chloroplast gene transcription. Bioessays 18, 465 471.
  • 24
    Sanger, F., Nicklen, S., Coulson, A.R. (1977) DNA sequencing with chain-terminating inhibitors. Proc. Natl Acad. Sci. USA 74, 5463 5467.
  • 25
    Reiss, T. & Link, G. (1985) Characterization of transcriptionally active DNA-protein complexes from chloroplasts and etioplasts of mustard (Sinapis alba L.). Eur. J. Biochem. 148, 207 212.
  • 26
    Fried, M. & Crothers, D.M. (1981) Equilibria and kinetics of lac repressor–operator interactions by polyacrylamide gel electrophoresis. Nucleic Acids Res. 9, 6505 6525.
  • 27
    Bülow, S. & Link, G. (1988) Sigma-like activity from mustard (Sinapis alba L.) chloroplasts conferring DNA-binding and transcription specificity to E. coli core RNA polymerase. Plant Mol. Biol. 10, 349 357.
  • 28
    Laemmli, U.K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680 685.
  • 29
    Merril, C.R., Goldman, D., Van Keuren, M.R. (1983) Silver staining methods for polyacrylamide gel electrophoresis. Methods Enzymol. 96, 230 239.
  • 30
    Meyer, H.E., Hoffmann-Posorske, E., Donella-Deana, A., Korte, H. (1991) Sequence analysis of phosphotyrosine-containing peptides. Methods Enzymol. 201, 206 224.
  • 31
    Fenyo, D., Qin, J., Chait, B.T. (1998) Protein identification using mass spectrometric information. Electrophoresis 19, 998 1005.
  • 32
    Baginsky, S., Tiller, K., Pfannschmidt, T., Link, G. (1999) PTK, the chloroplast RNA polymerase-associated protein kinase from mustard (Sinapis alba), mediates redox control of plastid in vitro transcription. Plant Mol. Biol. 39, 1013 1023.
  • 33
    Baginsky, S., Tiller, K., Link, G. (1997) Transcription factor phosphorylation by a protein kinase associated with chloroplast RNA polymerase from mustard (Sinapis alba). Plant Mol. Biol. 34, 181 189.
  • 34
    Gavel, Y. & Von Heijne, G. (1990) A conserved cleavage-site motif in chloroplast transit peptides. FEBS Lett. 261, 455 458.
  • 35
    Cline, K. & Henry, R. (1996) Import and routing of nucleus-encoded chloroplast proteins. Annu. Rev. Cell Biol. 12, 1 26.
  • 36
    Yang, J.J., Schuster, G., Stern, D.B. (1996) CSP41, a sequence-specific chloroplast mRNA binding protein, is an endoribonuclease. Plant Cell 8, 1409 1420.
  • 37
    Hu, J. & Bogorad, L. (1990) Maize chloroplast RNA polymerase: the 180-, 120-, and 38-kilodalton polypeptides are encoded in chloroplast genes. Proc. Natl Acad. Sci. USA 87, 1531 1535.
  • 38
    Hu, J., Troxler, R.F., Bogorad, L. (1991) Maize chloroplast RNA polymerase: the 78-kilodalton polypeptide is encoded by the plastid rpoC1 gene. Nucleic Acids Res. 19, 3431 3434.
  • 39
    Weihe, A., Hedtke, B., Börner, T. (1997) Cloning and characterization of a cDNA encoding a bacteriophage-type RNA polymerase from the higher plant Chenopodium album. Nucleic Acids Res. 25, 2319 2325.
  • 40
    Bogorad, L. (1991) Replication and transcription of plastid DNA. In The Molecular Biology of Plastids (Bogorad, L. & Vasil, I.K., eds), pp. 93 124. Academic Press, San Diego, USA.
  • 41
    Gruissem, W. & Tonkyn, J.C. (1993) Control mechanisms of plastid gene expression. Crit. Rev. Plant Sci. 12, 19 55.
  • 42
    Liu, K. & Hanna, M.M. (1995) NusA interferes with interactions between the nascent RNA and the C-terminal domain of the alpha subunit of RNA polymerase in Escherichia coli transcription complexes. Proc. Natl Acad. Sci. USA 92, 5012 5016.
  • 43
    Dantonel, J.C., Murthy, K.G.K., Manley, J.L., Tora, L. (1997) Transcription factor TFIID recruits factor CPSF for formation of 3′ end of mRNA. Nature 389, 399 402.
  • 44
    McCracken, S., Fong, N., Yankulov, K., Ballantyne, S., Pan, G.H., Greenblatt, J., Patterson, S.D., Wickens, M., Bentley, D.L. (1997) The C-terminal domain of RNA polymerase II couples mRNA processing to transcription. Nature 385, 357 361.
  • 45
    Barber, J. & Andersson, B. (1992) Too much of a good thing: light can be bad for photosynthesis. Trends Biochem. Sci. 17, 61 66.
  • 46
    Noctor, G. & Foyer, C.H. (1998) Ascorbate and glutathione: keeping active oxygen under control. Annu. Rev. Plant Physiol. Plant Mol. Biol. 49, 249 279.
  • 47
    Bowler, C., Van Montagu, M., Inzé, D. (1992) Superoxide dismutase and stress tolerance. Annu. Rev. Plant Physiol. Plant Mol. Biol. 43, 83 116.
  • 48
    Arisi, A.-C.M., Cornic, G., Jouanin, L., Foyer, C.H. (1998) Overexpression of iron superoxide dismutase in transformed poplar modifies the regulation of photosynthesis at low CO2 partial pressures or following exposure to the prooxidant herbicide methyl viologen. Plant Physiol. 117, 565 574.
  • 49
    Clark, G.B. & Roux, S.J. (1995) Annexins of plant cells. Plant Physiol. 109, 1133 1139.
  • 50
    Gidrol, X., Sabelli, P.A., Fern, Y.S., Kush, A.K. (1996) Annexin-like protein from Arabidopsis thaliana rescues DeltaoxyR mutant of Escherichia coli from H2O2 stress. Proc. Natl Acad. Sci. USA 93, 11268 11273.
  • 51
    Hausladen, A., Privalle, C.T., Keng, T., DeAngelo, J., Stamler, J.S. (1996) Nitrosative stress: activation of the transcription factor OxyR. Cell 86, 719 729.
  • 52
    Åslund, F., Zheng, M., Beckwith, J., Storz, G. (1999) Regulation of the OxyR transcription factor by hydrogen peroxide and the cellular thiol – disulfide status. Proc. Natl Acad. Sci. USA 96, 6161 6165.
  • 53
    Link, G., Tiller, K., Baginsky, S. (1997) Glutathione, a regulator of chloroplast transcription. In Regulation of Enzymatic Systems Detoxifying Xenobiotics in Plants (Hatzios, K.K., ed.), pp. 125 137. Kluwer Academic Publishers, Dordrecht, the Netherlands.
  • 54
    Lakhani, S., Khanna, N.C., Tewari, K.K. (1992) Two distinct transcriptional activities of pea (Pisum sativum) chloroplasts share immunochemically related functional polypeptides. Biochem. J. 286, 833 841.
  • 55
    Satoh, J., Baba, K., Nakahira, Y., Tsunoyama, Y., Shiina, T., Toyoshima, Y. (1999) Developmental stage-specific multi-subunit plastid RNA polymerase (PEP) in wheat. Plant J. 18, 407 416.DOI: 10.1046/j.1365-313x.1999.00465.x
  • 56
    Ohme, M., Tanaka, M., Chunwongse, J., Shinozaki, K., Sugiura, M. (1986) A tobacco chloroplast DNA sequence possibly coding for a polypeptide similar to E. coli RNA polymerase beta-subunit. FEBS Lett. 200, 87 90.
  • 57
    Hudson, G.S., Holton, T.A., Whitfeld, P.R., Bottomley, W. (1988) Spinach chloroplast rpoBC genes encode three subunits of the chloroplast RNA polymerase. J. Mol. Biol. 200, 639 654.
  • 58
    Igloi, G.L., Meinke, A., Döry, I., Kössel, H. (1990) Nucleotide sequence of the maize chloroplast rpo B/C1/C2 operon: Comparison between the derived protein primary structures from various organisms with respect to functional domains. Mol. Gen. Genet. 221, 379 394.
  • 59
    Hiratsuka, J., Shimada, H., Whittier, R., Ishibashi, T., Sakamoto, M., Mori, M., Kondo, C., Honji, Y., Sun, C.-R., Meng, B.-Y., Li, Y.-Q., Kanno, A., Nishizawa, Y., Hirai, A., Shinozaki, K., Sugiura, M. (1989) The complete sequence of the rice (Oryza sativa) chloroplast genome: intermolecular recombination between distinct tRNA genes accounts for a major plastid DNA inversion during the evolution of the cereals. Mol. Gen. Genet. 217, 185 194.
  • 60
    Amasino, R.M. & Crowell, D.N. (1991) Nucleotide sequence of an iron superoxide dismutase complementary DNA from soybean. Plant Physiol. 96, 1393 1394.
  1. Enzyme: DNA-dependent RNA polymerase (EC

  2. * Present address: Plant Physiology Department, University of Jena, D-07743 Jena, Germany.

  3. Present address: Department of Plant and Microbial Biology, UC Berkeley, CA, USA.