In vitro promoter recognition by the catalytic subunit of plant phage-type RNA polymerases

Abstract

We identified sequence motifs, which enhance or reduce the ability of the Arabidopsis phage-type RNA polymerases RPOTm (mitochondrial RNAP), RPOTp (plastidial RNAP), and RPOTmp (active in both organelles) to recognize their promoters in vitro with help of a 'specificity loop'. The importance of this data for the evolution and function of the organellar RNA polymerases is discussed. The single-subunit RNA polymerase (RNAP) of bacteriophage T7 is able to perform all steps of transcription without additional transcription factors. Dicotyledonous plants possess three phage-type RNAPs, RPOTm-the mitochondrial RNAP, RPOTp-the plastidial RNAP, and RPOTmp-an RNAP active in both organelles. RPOTm and RPOTp, like the T7 polymerase, are able to recognize promoters, while RPOTmp displays no significant promoter specificity in vitro. To find out which promoter motifs are crucial for recognition by the polymerases we performed in vitro transcription assays with recombinant Arabidopsis RPOTm and RPOTp enzymes. By comparing different truncated and mutagenized promoter constructs, we observed the same minimal promoter sequence supposed to be needed in vivo for transcription initiation. Moreover, we identified elements of core and flanking sequences, which are of critical importance for promoter recognition and activity in vitro. We further intended to reveal why RPOTmp does not efficiently recognize promoters in vitro and if promoter recognition is based on a structurally defined specificity loop of the plant enzymes as described for the yeast and T7 RNAPs. Interestingly, the exchange of only three amino acids within the putative specificity loop of RPOTmp enabled the enzyme for specific promoter transcription in vitro. Thus, also in plant phage-type RNAPs the specificity loop is engaged in promoter recognition. The results are discussed with respect to their relevance for transcription in organello and to the evolution of RPOT enzymes including the divergence of their functions.

Keywords: Arabidopsis; Chloroplast RNA polymerase; Mitochondrial RNA polymerase; Phage-type RNA polymerase; Promoter recognition.

Fig. 1

Deletion analysis of the mitochondrial Patp8-228/226 promoter. a Scheme of vectors used for in vitro transcription of the inserted Patp8-228/226 promoter sequences. The vectors contain ~200 bp (1, pKL23-Patp8-A), sequences from −27 to +5 (2, pKL23-Patp8-B) or −15 to +4 (3, pKL23-Patp8-C) around the in vivo transcription initiation sites at −228/226 determined previously by Kühn et al. (2005). For simplification, the scarcely detectable promoter Patp8-157 on vector pKL23-Patp8-A is not shown. Promoter cores are written bold, transcription initiation sites are underlined. Transcripts expected from initiation at the Patp8-228/226 and termination at hisa or thra (arrowheads) are indicated by horizontal black arrows labelled with the respective RNA lengths. b Recombinant RPOTm and RPOTp were assayed for promoter-specific transcription from supercoiled vectors (1–3) shown in (a). ³²P-labeled RNA products were separated in 5 % sequencing gels alongside an RNA size marker; sizes are given in nucleotides (marker lane not displayed). Specific RNA products are indicated at the right and labelled with the corresponding vector number. Additional minor signals have been observed before and may be due to differently migrating major products (Kühn et al. 2007). c 5′-RACE was performed on RNAs synthesized from pKL23-Patp8-B by RPOTm as described by Kühn et al. (2009). Transcripts were 5′ ligated to an RNA linker (+L) and subjected to RT-PCR. Non-ligated transcripts served as control (−L). Resulting PCR products were separated on an agarose gel alongside a size marker; sizes are given in base pairs (marker lane not displayed). The specific product corresponding to transcript 5′ ends mapping to Patp8-228/226 (arrowhead) was subjected to sequencing. The resulting chromatograms demonstrating the RNA linker and ligated transcript 5′ ends are shown below. Determined in vitro transcription initiation sites are indicated by bent arrows. Number of sequenced clones as well as frequency of initiation at the respective nucleotide are given within the chromatogram

Fig. 2

The promoter region of −27 to +6 is sufficient for specific transcription by RPOTm and RPOTp. a For construction of pKL23-Patp6-2-B 1 pKL23-Pycf1-B 2 pKL23-Prrn18-B 3 pKL23-Patp6-1-B 4, and pKL23-Patp6-1-C 5 indicated promoter sequences from −26 to +6 (or −27 to +5 in case of construct five) overlapping the transcription initiation site were inserted into pKL23. Transcripts expected from initiation at the introduced promoters followed by termination at hisa and thra are indicated. Symbols are as in Fig. 1. The motifs TATA and AGAG which are recurring in many mitochondrial and some plastidial NEP promoters, are highlighted in light and dark grey, respectively. b RPOTm and RPOTp were tested for promoter-specific transcription from supercoiled vectors (1–5) depicted in a). Transcript analysis and symbols are as in Fig. 1

Fig. 3

In vitro transcription from mutagenized promoter core motifs. Recombinant RPOTm and RPOTp were assayed for promoter-specific transcription of the mutagenized mitochondrial Patp8-228/226 (a) and Patp6-2-436 (b) promoters. Upper panels show wild-type and mutagenized promoter sequences which were inserted into pKL23 (compare Figs. 1a, 2b). Nucleotides in boldface highlight the promoter core motif targeted by the mutagenesis. The transcription initiation sites are underlined in the wild-type sequence. Transcripts expected from initiation at the introduced promoter followed by termination at hisa and thra have a length of 155 and 225 nt, respectively. Lane numbers in the lower panels correspond to promoter sequences shown above. Transcript analysis and symbols are as described in Fig. 1. Relative-fold changes for specific transcripts given below the autoradiogram were determined from the mean of two independent experiments

Fig. 4

Mutational analysis of the TATA and AGAG promoter motifs. Recombinant RPOTm and RPOTp were assayed for promoter-specific transcription of the chimeric mitochondrial Patp8-228/226 (a) and Patp6-1-156 (b) promoters. Upper panels display wild-type and mutagenized promoter sequences which were inserted into pKL23 (compare Figs. 1a, 2a). The TATA and AGAG promoter motifs targeted by the mutagenesis are highlighted in grey. Nucleotides exchanged between the Patp8-228/226 and Patp6-1-156 promoters are written in lower case letters. Lane numbers in the lower panels correspond to promoter sequences shown above. Transcript analysis and symbols are as described in Fig. 1. Relative-fold changes for specific transcripts given below the autoradiogram were determined from the mean of two independent experiments

Fig. 5

Mutational analysis of the AGAG motif in the mitochondrial Patp1-1898 and PtrnM-98 promoters. Recombinant RPOTm and RPOTp were assayed for promoter-specific transcription of the mutagenized Patp1-1898 (a) and PtrnM-98 (b) promoters. Upper panels display wild-type and mutagenized promoter sequences which were inserted into pKL23. The AGAG promoter motif targeted by the mutagenesis is highlighted in grey. Mutagenized nucleotides are written in lower case letters. Lane numbers in the lower panels correspond to promoter sequences shown above. Transcripts expected from initiation at the introduced promoter followed by termination at hisa and thra have a length of 159 and 228 nt, respectively. Transcript analysis and symbols are as described in Fig. 1. Relative-fold changes for specific transcripts given below the autoradiogram were determined from the mean of two (Patp1-1898) or three (PtrnM-98) independent experiments

Fig. 6

Mutagenesis of RPOTmp. a Amino acid sequence alignment of the putative specificity loop regions of the phage-type RNAPs from A. thaliana (At-RPOTmp, At-RPOTm, and At-RPOTp) and the mitochondrial phage-type RNAP from yeast (Sc-RPO41). The specificity loop region as well as the conserved G + H and I-blocks defined from sequence similarity of single subunit RNAP amino acid sequences by Cermakian et al. (1997) are displayed by GeneDoc software (http://external.informer.com/nrbsc.org/gfx%2Fgenedoc). Shaded positions are conserved in 75 % (grey) or 100 % (black) of aligned sequences. Amino acids that were exchanged in RPOTmp are boxed (RPOTmp: T884R, K885H, H898R). Amino acids of the yeast enzyme described to directly interact with bases of the promoter sequence are underlined (Nayak et al. 2009). b Coomassie-stained polyacrylamide gels demonstrating purification of recombinant RPOTmp wild-type (Tmp) and mutagenized proteins (Tmp (RHR)), respectively. Histidine-tagged proteins were expressed in E. coli by the pCOLD system, purified from bacterial extracts over Ni²⁺-agarose and separated by SDS–PAGE. Shown are total proteins of uninduced bacterial cultures (U) alongside the purified recombinant enzymes (P) which are designated by arrow heads. Sizes of the molecular mass marker are indicated in kilodaltons. The expected molecular weight for RPOTmp/RPOTmp (RHR) is 106 kDa. c Unspecific in vitro transcription by mutagenized RPOTmp. Incorporation of [α−³²P]-UMP into transcripts synthesized in vitro by recombinant RPOTmp wild-type (Tmp WT) and mutagenized protein (Tmp(RHR)) from calf thymus DNA was determined using scintillation counting. 100 % transcriptional activity corresponds to a complete incorporation of ³²P UTP into the RNA transcripts. As a positive control served the T7 RNA polymerase (T7). Error bars represent standard deviations from three independent experiments

Fig. 7

Specific in vitro transcription of Patp8-228/226 and PtrnM-98 by mutagenized RPOTmp. a wild-type (WT) and mutagenized RPOTmp (RHR) were assayed for promoter-specific transcription from pKL23-Patp8-A (compare Fig. 1a). b Wild-type and mutagenized RPOTmp were assayed for promoter-specific transcription from pKL23-PtrnM-98 (lanes 1) and pKL23-PtrnM-98 tata (lanes 2) (compare Fig. 5b). Transcript analysis and symbols are as in Fig. 1. Relative-fold changes for specific transcripts given below the autoradiogram were determined from the mean of two independent experiments. Transcripts in (b) were separated on the same gel but not immediately adjacent to each other which is indicated by grey vertical lines

Fig. 8

Organellar transcription initiation by phage-type RNAPs. Depending on the individual promoter there seem to be at least three modes of promoter recognition conceivable. While some promoters might be directly recognized by the intrinsic capacity of RPOTm/Tp after the promoter region has been opened by the RNAP alone (a with supercoiled DNA templates as a precondition) or with support of one or more hypothetical DNA-binding protein(s) (DBP) (b), others are likely to additionally require one or more associated and yet unidentified specificity factor(s) (SF) mediating the recognition of the promoter sequence and/or guiding the RNAP to the transcription start site (c)