|
|
|
|||||||
|
|||||||||
(Received for publication, April 24, 1995; and in revised form, June 21,
1995) From the
To elucidate the mechanism involved in the transcription
initiation process in mitochondria of dicotyledonous plants, an in
vitro transcription system was established for pea (Pisum
sativum L.). The partially purified mitochondrial protein extract
initiates transcription on homologous pea templates as well as on
heterologous mitochondrial DNA from other dicot plant species. In
vitro transcription begins within the nonanucleotide
5`-
The resident mitochondrial genomes encode essential genes, whose
expression is indispensable for the function of the mitochondria and
thus for the survival of the eukaryotic cell (Levings and Brown,
1989). Towards understanding the regulatory control mechanisms involved
in mitochondrial gene expression, the biochemistry of transcription of
the mitochondrial DNA has been intensively studied in animals and
fungi. The mammalian 16-kilobase mitochondrial genome is symmetrically
transcribed from two major promoters, one for each of the two different
DNA strands. Transcription of the Saccharomyces cerevisiae mitochondrial DNA is initiated at about 20 copies of a highly
conserved promoter motif (Christianson and Rabinowitz, 1983). In
vitro transcription studies showed that in both yeast and mammalia
at least two protein components are engaged in specific and efficient
transcription initiation. Although in yeast all genes and the encoded
proteins of the mitochondrial transcription machinery are described in
detail, only the gene for the transcription factor h-mtTFA has been
identified in human mitochondria (Shadel and Clayton, 1993). Compared with animals and fungi the mode of transcription initiation
in mitochondria of plants remains elusive. Their much larger and more
complex mitochondrial genomes and the frequently observed complex
transcription patterns of individual genes impede the identification
and analysis of promoters in mitochondria of plants (Levings and Brown,
1989; Newton, 1988). Several transcription initiation sites have
clearly been identified by in vitro capping analyses in
mitochondria from both monocot and dicot plant species. Inspection of
sequences at these potential promoter regions identified distinct
sequence motifs to which promoter function has been attributed
(Mulligan et al., 1988a, 1988b, 1991; Covello and Gray, 1991;
Brown et al., 1991; Binder and Brennicke, 1993a, 1993b). The successful development of in vitro transcription
systems for the monocot plant species wheat and maize confirms the
competence of several promoters of protein coding genes in vitro (Hanic-Joyce and Gray, 1991; Rapp and Stern, 1992). Detailed
investigation by deletion analysis, linker-scanning mutagenesis, and
site-directed mutagenesis revealed the atp1 promoter in maize
to be composed of a central domain extending from -7 to +5
and a 3-base pair upstream domain located between positions -10
and -12. The central element contains the highly conserved
5`-YRTA-3` core element that is consistent with the 5`-CRTA-3` motif
defined by sequence inspection at the transcription initiation sites
(Rapp et al., 1993; Mulligan et al., 1991). This
motif is observed at most of the identified monocot mitochondrial
promoters, which otherwise show only limited similarity between
different species (Mulligan et al., 1991; Gray et
al., 1992). In dicot plant mitochondria, in vitro capping analysis of primary transcripts indicates a much better
conservation of most of the potential promoter sequences. A conserved
nonanucleotide motif
5`- To extend our knowledge about the mode of transcription initiation
in dicot plant mitochondria we have now established an in vitro transcription system for pea. The in vitro results
indicate the conserved nonanucleotide motif to be essential for
transcription initiation and confirm its function as a general promoter
motif in different dicot plant species. In addition to mRNA promoters
the pea mitochondrial extract initiates transcription at a tRNA
promoter suggesting that the transcription activity in this in
vitro system is not restricted to protein coding genes.
The soybean mtDNA clones KM57BK3
and KM2F2EH containing transcription initiation sites for the atp9 and RNA b/c gene, respectively, were kindly
provided by Dr. Gregory Brown (Brown et al., 1991). Clone
satp9XR482 is a 0.48-kilobase XbaI/RsaI sublone of
KM57BK3. Clone opheBH500 contains the transcription initiation site
identified upstream of the Oenothera tRNA All mtDNA fragments were cloned into pBluescript (Stratagene)
vectors using standard cloning techniques (Sambrook et al.,
1989). For in vitro transcription reactions plasmid DNA was
purified by centrifugation on CsCl gradients and linearized with
restriction enzymes. After digestion, template DNA was checked on
agarose gels for complete cleavage, extracted with phenol/chloroform,
precipitated with ethanol, and resuspended in double distilled water.
DNA concentrations were determined photometrically by absorption at 260
nm.
The search for a suitable dicot plant species and tissue
type was finally successful in identifying pea seedlings as a suitable
source for reasonable quantities of transcriptionally active
mitochondria. Because no in vitro capping analysis had been
carried out with pea transcripts, only little data were available on
transcription initiation sites in mitochondria of this plant species.
Homologous transcription assays thus relied on a DNA fragment
containing the most distal 5` mRNA terminus mapped in the upstream
region of the atp9 gene (Fig. 1A, terminus (1) (Morikami and Nakamura, 1993). The sequence
surrounding this 5` end
(5`-
Figure 1:
Enrichment procedure for promoter
specific in vitro transcription activity in pea mitochondrial
extracts. A, structure of the genomic fragment containing the atp9 coding region present in the type I allele in pea
mitochondria (Morikami and Nakamura, 1987, 1993). Three 5` ends ((1)-(3), indicated by bent arrows) have been mapped
upstream of the coding region. A SfuI/KpnI fragment
containing the most distal 5` end ((1)) located 1048 nt
upstream of the start codon was subcloned (patp9SK630) and
used for in vitro transcription reactions. A 355-nt-long
run-off transcript (dotted arrow) is expected for correct
transcription initiation on the KpnI-linearized DNA template.
The sequence from -7 to +2 surrounding this 5` end (underlined) shows high similarity to the putative promoter
consensus sequence from mitochondria of dicot plants (Brown et
al., 1991; Binder and Brennicke, 1993a, 1993b). Restriction sites
are indicated for ClaI (C), KpnI (K), and SfuI (Sf). B, gel analysis
of in vitro transcription assays with different fractions of
the pea mitochondrial protein extracts and KpnI-linearized
patp9SK630 DNA templates. Fractions used in the assays are: lysed
mitochondria (mitochondria), S100 supernatant (S100),
20% and 50% ammonium sulfate fractions (20% AS and 50% AS), and fractions eluted from the MonoQ column with 50, 100, 200,
250, 300, 400, 500 and 750 mM KCl. Sizes of DNA length
standards are given in nucleotides.
No specific transcription products were observed in
assays with only vector sequences or without any DNA template,
indicating their dependence upon mitochondrial promoter sequences. The
absence of any transcription product in a reaction carried out in the
presence of RNase A showed the products of these assays to be indeed
RNA resulting from genuine transcription events (data not shown).
Because the majority of the specific transcription initiation activity
eluted at 250 mM KCl, this fraction was used for the detailed
investigation of the accuracy of the initiation process in vitro and for heterologous transcription assays. To confirm that
transcription is indeed initiated within the mitochondrial insert, two
differently linearized pea atp9 templates (patp9SK630 and
patp9SC550, see ``Material and Methods'') were tested in the
transcription assays. In all instances run-off transcripts of the
predicted sizes were observed in these reactions, and the lengths of
the in vitro synthesized RNAs differed by the distances of the
respective restriction sites used for linearization of the DNA
templates confirming correct initiation within the mitochondrial insert
(data not shown).
Figure 2:
The soybean atp9 promoter is
recognized by the pea mitochondrial in vitro transcription
system. A, a transcription initiation site (marked by a bent arrow) has been mapped by in vitro capping
analysis upstream of the soybean atp9 gene (Brown et
al., 1991). The 0.5-kilobase XbaI/RsaI and
1.69-kilobase BamHI/KpnI fragments, both containing
this promoter, were subcloned (satp9XR482 and KM57BK3) and used for in vitro transcription reactions with the 250 mM KCl
fraction of the pea mitochondrial extracts. Sizes of the run-off
products (dotted lines) expected from transcription of KpnI-linearized DNA templates are 280 and 238 nt,
respectively. Vector sequences are indicated by dashed lines;
restriction sites are given for BamHI (B), XbaI (X), KpnI (K), and RsaI (R). B, gel analysis of run-off
transcripts transcribed from the DNA templates as shown above.
Numbering of the gel lanes correlates with the expected run-off
transcripts. Signals corresponding to run-off transcripts of the
predicted sizes are observed in both reactions. DNA length standards
are given in nucleotides.
In
addition to the atp9 promoter clone, KM2F2EH carrying two
transcription initiation sites for RNAs of unidentified function (RNA b, c, and e) (Brown et al., 1991)
was subjected to heterologous assays (Fig. 3). Although the
common transcription initiation site for RNA b and c conforms to the consensus sequence element, sequences at the
transcription start site for RNA e show no similarity to this
promoter motif (Brown et al., 1991). Transcripts with the
sizes predicted for correct initiation at the promoter for RNA b and c were observed with two differently linearized
templates (Fig. 3). However, no transcripts of the respective
sizes were observed for RNA e, suggesting that this promoter
is not recognized by the pea extract.
Figure 3:
Specificity of the pea mitochondrial
extract for promoters containing the conserved nonanucleotide motif. A, the genomic location containing two transcription start
points (bent arrows) for three different RNAs in soybean
mitochondria (Brown et al., 1991). The initiation sites are
indicated by bent arrows and designated (b/c) and (e) corresponding to the transcribed RNAs b/c and e. Transcription initiation at start point (b/c) should yield run-off transcripts of 330 and 298 nt (dotted
arrows) on an EcoRI/HindIII clone (KM2F2EH) linearized with KpnI (site located in the
vector sequence) and HindIII, respectively. Restriction sites
are indicated for EcoRI (E), HindIII (H), and KpnI (K). B, run-off
transcripts checked by polyacrylamide gel electrophoresis show correct
transcription initiation at site (b/c) in the in vitro transcription reactions. Lanes with the transcription assays are
numbered corresponding to the expected run-off transcripts shown in A. Transcripts of the sizes expected from transcription
initiation at site (e) are not observed in these assays. Sizes
of coelectrophoresed DNA length standards are given in
nucleotides.
To extend the analysis of
heterologous transcription initiation, cox2 and atp1 templates from Oenothera were tested with the pea
extract. In these assays distinct albeit weak signals corresponding to
run-off transcripts of the predicted sizes showed correct recognition
of these heterologous templates by the pea extract (data not shown).
Figure 4:
A promoter located upstream of a
tRNA
In the
analysis of homologous pea atp9 transcripts, the 5` ends of
the in vitro transcribed RNAs are scattered around the
transcription start site used in vivo. In addition to the 5`
end of the correctly initiated transcript corresponding to the primer
extension length obtained on in vivo transcripts (data not
shown), signals corresponding to RNAs up to eight nucleotides longer
were detected in this analysis (Fig. 5A). Single
signals were obtained in the analysis of 5` ends of RNAs derived from
heterologous transcription assays. An extension reaction with RNAs
synthesized in vitro from a soybean atp9 template
(satp9XR482) detects the guanosine, which is also the first transcribed
nucleotide in vivo (Fig. 5B). Signals
corresponding to adenosine and guanosine at positions -2 and
-1, respectively, are detected in the analysis of in vitro soybean RNA b/c transcripts (Fig. 5C). The difference of 1 or 2 nucleotides in
comparison to the 5` end mapped for in vivo transcripts might
be due to a slightly altered migration behavior of the cDNA molecule
compared with the sequencing products.
Figure 5:
Primer extension analysis of in vitro synthesized RNA derived from transcription reactions using pea
mitochondrial protein extract and homologous and heterologous DNA
templates. The primer extension reaction was performed on in vitro transcripts recovered from a polyacrylamide gel. To determine the
exact 5` ends of the run-off transcripts the extension products (pex) were coelectrophoresed with sequencing reactions
performed with the same oligonucleotides as in the primer extension
reactions. For easier interpretation, the sequencing reactions have
been labeled in the inverted order CTAG to show the sequence of the
sense strand given in the right margins. Signals of the
extension reactions are indicated by horizontal arrows; 5`
ends determined from isolated in vivo transcribed RNAs are
indicated by bent arrows (Brown et al., 1991;
Morikami and Nakamura, 1993). The boxed sequence represents
the conserved nonanucleotide sequence element of mitochondria from
dicotyledonous plants. A, primer extension analysis of 5` ends
of in vitro transcripts derived from KpnI-linearized
DNA template patp9SK630 containing the pea atp9 promoter. B, analysis of in vitro transcription products
obtained with KpnI-linearized soybean atp9 template
satp9XR482. C, determination of the 5` ends of run-off
transcription products obtained in transcription reactions with KpnI-digested RNA b/c template (KM2F2EH)
from soybean mitochondria.
Despite the ambiguous results
obtained in the investigation of the pea atp9 transcripts, the
primer extension analysis of heterologous in vitro transcripts
derived from soybean templates indicates the capability of the pea
mitochondrial lysate to accurately initiate transcription at the start
sites determined for in vivo transcribed RNA.
Figure 6:
Deletion analysis of the nontranscribed
region upstream of the pea atp9 transcription start site. A, using PCR techniques, sequences were successively removed
from clone patp9SK630. Designations of the constructs used in the
transcription assays are given in the left margin. Bold
lines illustrate mitochondrial sequences. Vector sequences are
given as thin lines. Numbers indicate nucleotide
positions with respect to the transcription start site (+1, shown
as bent arrows). The conserved nonanucleotide motifs are
indicated by black boxes. Restriction sites are indicated for KpnI (K), SfuI (Sf), and BamHI (B). B, part of a PhosphorImager print
of the transcription products separated on a polyacrylamide gel. The
size of the run-off products obtained upon in vitro transcription of KpnI-linearized templates (designations
given in the top part) is shown in the left margin.
Sizes of coelectrophoresed length standards (M) are shown at
the right hand side. Faint transcription products are observed
with
Fractionating pea
mitochondrial extracts on a MonoQ column showed that the formation of
transcription initiation complexes is favored by the enrichment of RNA
polymerase and potential transcription factors in an optimal
stoichiometric proportion in the 250 mM KCl elution step. The
vast majority of nonspecific transcription activity (i.e. RNA
polymerase separated from specificity factors) elutes with the 400
mM KCl fraction. Nonspecific transcription activity is
additionally enhanced by the contamination of this fraction with
endogenous mtDNA, which elutes from the MonoQ column at KCl
concentrations higher than 300 mM. Cofractionating intact
mtDNA-RNA polymerase complexes possibly also contribute to background
activity. High molecular weight transcripts detected in all
transcription assays are most likely due to nonspecific initiation
events. Similar transcripts were also observed in other in vitro transcription systems and are attributed to initiation events at
ends or nicks of the linear DNA templates (Hanic-Joyce and Gray, 1991;
Rapp and Stern, 1992). Further purification of the active protein
fractions supplemented by DNA binding assays will be necessary to
characterize the individual protein components involved in the
transcription initiation process.
In the pea system described here, transcription is
indeed initiated precisely on the soybean atp9 template (Fig. 5B). With the soybean RNA b/c template, a 5` end is detected that is very close to the 5`
transcript terminus determined by an in vitro capping analysis (Fig. 5C). We assume that the differences between 5`
ends of in vitro run-off transcripts, which are scattered over
several nucleotides, and of in vivo pea atp9 mRNAs,
where a single G is detected as a 5` terminal nucleotide, are derived
through intrinsic problems of the primer extension experiments (Fig. 5A). The in vitro synthesized pea atp9 transcripts, which are recovered from the gel and used as
RNA templates in the extension analysis, are always detected as a
single sharp signal as seen for example in Fig. 1B. A
series of transcripts with different 5` ends ranging over 10
nucleotides, as indicated by the primer extension analysis, should
rather appear as an expanded signal separated in a 5% polyacrylamide
gel (Fig. 1B). Because a single signal consistent
with the result obtained by Morikami and Nakamura(1993) (indicated by a bent arrow in Fig. 5A) is always detected in
primer extension experiments with isolated in vivo pea mtRNA
(data not shown), we conclude that the artificial length and
composition of the in vitro run-off transcripts might be
responsible for the scattered termini in the primer extension
experiments most likely by a disadvantageous back folding.
A comparison of sequences
located between nucleotide positions -25 and -8 reveals a
high content of adenosine-thymidine base pairs (67-83%) with
almost 100% of the AT base pairs concentrated between nucleotide
positions -13 to -9. This AT-box and a conserved adenosine
nucleotide at position -are the most remarkable features within
the critical upstream region (Fig. 7). The presence of the
AT-rich region is most likely responsible for enhanced melting of the
DNA strands during the transcription initiation process.
Figure 7:
Alignment of sequences containing the
promoters transcribed by the pea mitochondrial extract. Sequences
critical for the transcription of the pea atp9 promoter were
compared with other promoters recognized by the pea lysate and the
maize atp1 promoter, respectively (Rapp et al.,
1993). Nucleotide positions are given with respect to the first
transcribed nucleotide (position +1, underlined). Highly
conserved sequences in the dicot promoter regions are boxed and designated AT-box (for AT-rich box) and CNM (for conserved nonanucleotide motif). The 5`-YRTA-3` motif common
to dicot and monocot promoter sequences is underlined.
Sequence elements critical for the transcription of the maize atp1 promoter are the ``upstream domain'' (boxed)
and the core element (underlined) of the central domain.
Adenosines and thymidines are shown in bold and the
adenosine-thymidine contents of sequences between nucleotide position
-25 and -8 are given in the right margin. The
distances between the first transcribed nucleotide, the center of the
AT-box at position -10, and the conserved adenosine at -20
locate these elements to the same side of the DNA-helix, suggesting
their recognition by a protein factor(s). Although the conserved
nonanucleotide motif element is essential for transcription, the
AT-rich box is found to be required to raise transcription initiation
to significant levels.
This model
is similar to the maize atp1 promoter structure (Rapp et
al., 1993), although the core element in dicot plants appears to
be extended and much better conserved than their counterpart in
monocots. Additional detailed deletion and mutagenesis studies will
define the exact promoter structure and requirements in mitochondria of
dicot plant species.
Only a few of
these alternative promoters have been identified and so far no
similarity could be detected between their primary structures. These
nonconserved promoters represent probably gene- and/or species-specific
single copy promoters. Because in animal and yeast mitochondria
replication is primed by short RNAs initiated at promoter sequences, a
similar function might be attributed to some of these alternative
promoters in plants (Schmitt and Clayton, 1993; Xu and Clayton, 1995).
Much more data about these promoters in a single plant species are
necessary to see whether different types or classes of promoters are
indeed present in plant mitochondria and how they are recognized.
Volume 270,
Number 38,
Issue of September 22, pp. 22182-22189, 1995
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
FOOTNOTES
ACKNOWLEDGEMENTS
REFERENCES
CRTAAGAGA
-3` (transcription
start site is underlined) conserved at most of the identified
transcription initiation sites in dicot plant mitochondria. The in
vitro initiation at promoters of protein as well as of tRNA coding
genes indicates a common mode of transcription initiation for different
types of RNA. The competent recognition of different heterologous
templates supports a general functional role of the conserved
nonanucleotide within mitochondrial promoters of dicotyledonous plants.
Initial studies of the promoter structure by deletion analysis in the
5` region of the pea atp9 promoter show that in addition to
the conserved nonanucleotide, which is essential for transcription
initiation in vitro, sequences up to 25 nucleotides upstream
of the transcription start site are necessary for an efficient
initiation event.
CRTAAGAGA
-3` has been
defined by comparison of sequences surrounding mRMA, rRNA, and tRNA
transcription initiation sites. With monocot promoters, only the
tetranucleotide 5`-CRTA-3` appears to be conserved. Because the
nonanucleotide motif is common to transcription initiation sites of
different dicot plant species, it may represent a general core element
of mitochondrial promoters in this plant group (Brown et al.,
1991; Binder and Brennicke, 1993a, 1993b). However, several
transcription initiation sites have been determined that show no
sequence similarity to this conserved nonanucleotide motif or any other
known plant mitochondrial promoter. Structure and function of these
most likely gene- and/or species-specific potential promoters remain
unclear (Brown et al., 1991; Binder et al., 1994).
Recombinant DNA Templates
Plasmid clones
patp9SC550 and patp9SK630 contain sequences covering the most distal 5`
mRNA end of the atp9 gene subcloned from the type I allele of
the 4 different atp1/atp9 arrangements present in pea
mitochondrial DNA (Morikami and Nakamura, 1987, 1993). For deletion
analysis of the pea atp9 promoter region, DNA fragments were
amplified by PCR (
)using patp9SK630 as template and the
following oligonucleotides: clone 
patp9-67, dp-67
(5`-GTGGATCCTTATGTGAGGTTCTTTCC-3`) and dp+355
(5`-TGGGTACCTCATAGGGC-3`); clone patp9-35, dp-35
(5`-GTGGATCCTTGTTTTGAGTACTCGAC-3`) and dp+355; clone
patp9-25, dp-25 (5`-GTGGATCCTACTCGACGAAATAATAG-3`)
and dp+355; clone patp9-7, dp-7
(5`-GTGGATCCCATAATAAGAGAAGATATTGG-3`) and dp+355; and clone
patp9+2, dp+2 (5`-GTGGATCCAGATATTGGACAATTGAG-3`) and
dp+355. Designation of the upstream oligonucleotides reflects,
with respect to the transcription start site, the position of the most
5` nucleotide within the oligonucleotide, which is identical with the
mitochondrial sequence. A BamHI restriction site was
introduced into each upstream oligonucleotide for cloning. The
downstream oligonucleotide dp+355 covers a KpnI site
contained in the mitochondrial DNA sequence. Amplified DNA fragments
were digested with the respective restriction enzymes and cloned into
pBluescriptII KS(-) vectors. gene
on a DNA fragment obtained by PCR with oligonucleotides OP-1
(5`-GGAAATCCAAGGAGGTGGC-3`) and P7 (5`-ATAAGCTTGAATTTCCAAATCCGG-3`).
Preparation of Mitochondrial Protein Extracts
Pea
seedlings (Pisum sativum L., var. Progress No. 9 and var.
Lancet) were grown in the dark for 7 days. Mitochondria were isolated
by differential centrifugation and purification on Percoll gradients as
described previously (Binder and Brennicke, 1993a). The mitochondrial
protein extract was prepared basically following a method established
by Hanic-Joyce and Gray(1991) for wheat and modified by Rapp and
Stern(1992) for maize mitochondria. About 5 g of purified mitochondria
were lysed in the presence of 0.5% Triton X-100 and 1 M KCl.
Membranes were pelleted by centrifugation at 100,000 
g. The supernatant (S100) was diluted with 1 volume of buffer
V containing 90 mM Tris-HCl, pH 8.0, 0.1 mM EDTA, 0.5
mM dithiothreitol, and 2.5% glycerol, and hydrophobic proteins
were precipitated by the addition of 0.224 g of solid
(NH
)
SO/ml of diluted solution and
centrifugation at 16,000 g for 30 min (20% AS
fraction). The remaining protein was precipitated from the supernatant
by the addition of 0.41 g of solid
(NH)SO/ml of solution and
centrifugation at 16,000 g for 90 min (50% AS
fraction). The protein pellet was then resuspended in 1 ml of buffer A
(10 mM Tris-HCl pH 8.0, 1 mM dithiothreitol, 0.1
mM EDTA, and 7.5% glycerol) containing 50 mM KCl and
dialyzed for several hours against 3 liters of the same buffer. For
further purification by anion-exchange chromatography, this protein
fraction was applied to a MonoQ column (Pharmacia Biotech Inc.)
equilibrated with buffer A (with 50 mM KCl added). Bound
protein was eluted from the column with 100, 200, 250, 300, 400, 500,
and 750 mM KCl in buffer A. Fractions of each elution step
were pooled, dialyzed against 3 liters of buffer A containing 10 mM KCl, concentrated by ultrafiltration using Centricon 10 (Amicon),
rapidly frozen in liquid nitrogen, and stored at -80 °C.
Protein concentrations were measured using a Bio-Rad (Bradford) protein
assay.In Vitro Transcription Assays
The in vitro transcription reactions were performed in a total volume of 12.5
µl of reaction mixture containing 10 mM Tris-HCl, pH 7.9,
10 mM MgCl, 1 mM dithiothreitol, 20
mM KCl, 500 µM each of ATP, CTP, and GTP, 25
µM UTP, 40 units of RNase inhibitor (Boehringer Mannheim),
10 µCi of [
-P]UTP (3000 Ci/mmol), and
100-500 ng of linearized template DNA. Reactions were started by
the addition of 30-80 µg of the protein extract and were
incubated for 30 min at 30 °C. After the addition of 37 µl of
stop mix (4.8 M urea, 0.4 M sodium acetate, 5.3
mM aurintricarboxylic acid, 30 µg/ml tRNA (wheat germ),
and 0.8% (w/v) SDS), reaction mixes were extracted with
phenol/chloroform, and nucleic acids were precipitated with ethanol.
Total nucleic acids were resuspended in 4 µl of loading solution
(80% (v/v) formamide, 50 mM Tris borate, pH 8.3, 1 mM EDTA, 0.1% (w/v) xylene cyanol, and 0.1% (w/v) bromphenol blue)
and electrophoresed on 5% polyacrylamide gels, and transcription
products were examined by autoradiography.
Primer Extension Analysis
Primer extension
analysis was carried out with in vitro transcripts recovered
from the polyacrylamide gels in a buffer containing 500 mM
ammonium acetate, 0.1 mM EDTA, and 0.1% (w/v) SDS. RNA was
resuspended in 7 µl of double distilled water and mixed with 5
µl of 5` end-labeled primer (5 10
cpm). The
mixture was incubated for 10 min at 70 °C and for another 10 min at
42 °C. After the addition of 2 µl of 10 synthesis
buffer, 2 µl 4 mM dNTP mix and 2 µl of 0.1 mM dithiothreitol, the reaction was started with 1 µl of
Superscript reverse transcriptase (200 units/µl, Life
Technologies, Inc.) and incubated for 1 h at 42 °C. Reaction
products were precipitated with ethanol, resuspended in 4 µl of
loading solution, and analyzed by polyacrylamide gel electrophoresis.
Primer extension analysis of the in vitro transcripts of pea atp9 template patp9SK630 was performed using oligonucleotide
PA-14 (5`-CAAAGGAGGAACTCCCG-3`) complementary to sequences between
nucleotides +49 and +65 (first transcribed nucleotide is
+1). Oligonucleotides A9 (5`-GTGAACAGAAGCTTTCTCGG-3`) and SBC-1
(5`-CTGCACACCAAGCCAATCTCG-3`) complementary to sequences between
nucleotides +72 and +91 (atp9) and nucleotides
+71 and +91 (RNA b/c), respectively, were
used for the primer extension analysis of the soybean atp9 and
RNA b/c in vitro transcripts.
Development of an in Vitro Transcription System for
Dicot Plant Mitochondria
The recent identification of several
mitochondrial transcription initiation sites of soybean and Oenothera by in vitro capping analysis (Brown et
al., 1991; Binder and Brennicke, 1993a, 1993b) provided the DNA
templates required for the preparation of a transcriptionally active
mitochondrial extract. The isolation of such transcriptionally active
extracts from dicot plant mitochondria was additionally guided by the
successful development of an in vitro transcription system for
monocot plants (Hanic-Joyce and Gray, 1991; Rapp and Stern, 1992).
Because data on functional transcription initiation sites were only
available for Oenothera and soybean, the isolation of
transcriptionally active protein fractions was started in these two
plant species. It soon became evident, however, that the impossibility
of isolating large quantities of mitochondria from Oenothera or soybean and the high content of nucleases in protein extracts
of these plant species prohibited the large scale preparation of
transcriptionally active lysates. Attempts to prepare protein extracts
from large amounts of mitochondria isolated from potato tubers likewise
did not yield the desired activities, possibly owing to the basically
low transcriptional activity in resting tissues like tubers (data not
shown).CATAAGAGA
-3`), which was
determined by primer extension and S1 protection analysis, shows high
similarity with the conserved nonanucleotide element detected at
primary 5` transcript termini in mitochondria of other dicot plants
(Brown et al., 1991; Binder and Brennicke, 1993a, 1993b).
Specific Transcription Initiation by a Pea Mitochondrial
Protein Extract
The KpnI-linearized patp9SK630 template
containing the region described above (Fig. 1A) was
used in homologous in vitro transcription assays to test all
protein fractions obtained during the enrichment procedure (see
``Materials and Methods''). A weak signal corresponding to a
transcript of the predicted size (Fig. 1A, 355
nt) was detected in the assay performed with the 50% ammonium
sulfate fraction (Fig. 1B, 50% AS). The
additional fractionation of the 50% AS cut resulted in enrichment of
the specific transcription initiation activity. The majority of the
specific activity could be detected in the 250 mM KCl MonoQ
fraction represented by the signal in the size range expected for the
correct run-off transcript. Additional weaker transcription initiation
activity was also present in the 300 mM KCl MonoQ fraction, as
indicated by a faint band of the predicted size (Fig. 1B, 250 and 300 mM KCl).Transcription Is Initiated on Heterologous Templates from
Soybean
The conservation of the potential promoter sequences
identified by in vitro capping analysis of soybean and Oenothera transcripts and by the above in vitro studies of the pea atp9 gene indicates similar
transcription initiation complexes most likely including homologous
protein components within these dicot plant species. This hypothesis
was tested by heterologous transcription assays with mitochondrial
extracts and DNA templates from different dicot plant species. Clones
KM57BK3 and satp9XR482 containing the mitochondrial promoter of the
soybean atp9 gene were assayed for transcription initiation in
pea protein extracts (Fig. 2A). Both templates direct
transcription of RNAs corresponding to the predicted sizes, indicating
accurate initiation in these heterologous reactions (Fig. 2B, lanes 1 and 2).
In Vitro Transcription of a tRNA Gene
Promoter
In vitro capping analysis of primary
transcripts covering tRNA genes encoded in Oenothera mitochondria had identified a transcription initiation site
located upstream of three alleles of the gene for tRNA and a gene for
tRNA
. The sequence at this
transcription initiation site is consistent with the conserved sequence
element derived from protein coding and rRNA genes (Binder and
Brennicke, 1993b). To test whether this promoter is also recognized and
transcribed by the pea extract, clone opheBK500 containing a
PCR-amplified 500-nt-long DNA fragment with the complete gene for
tRNA
was used in the transcription analysis. A run-off
transcript of 301 nt is expected from correct transcription initiation
at this tRNA promoter and full-length elongation along the HindIII-cleaved template (Fig. 4A). Analysis
of the transcription products by polyacrylamide gel electrophoresis
shows a RNA of the expected size to be synthesized in the in vitro reaction (Fig. 4B). This result confirms common
features for promoters of both protein coding and tRNA genes.
gene in Oenothera mitochondria is
recognized in the pea in vitro transcription system. A, a cloned PCR product (opheBK500) containing a copy
of this gene and about 380 nt of the upstream region was used for in vitro transcription assays with the pea extract. The
correct recognition of this potential promoter (marked by a bent
arrow) on a HindIII-cleaved template should result in the
transcription of a 301-nt-long RNA (dotted arrow). A
restriction site is indicated for HindIII (H). S/Bl marks the blunt end of the DNA fragment cloned in the SmaI restriction site of the vector (dashed lines). B, gel analysis of in vitro transcription experiments
shows a run-off transcript of the predicted size. Sizes of
coelectrophoresed DNA length standards are given in
nucleotides.
Precise Mapping of 5` Ends of in Vitro Synthesized
RNAs
The accuracy of the in vitro transcription
initiation event was determined by primer extension analysis carried
out with specific transcripts synthesized in both homologous and
heterologous in vitro transcription reactions.
Deletion Analysis of the Pea atp9
Promoter
Analysis of the pea atp9 promoter focused on
the contribution of the region upstream of the transcription start site
and on the role of the conserved nonanucleotide as a functional element
in mitochondrial promoters of dicot plants. For this purpose a series
of constructs with progressively deleted sequences upstream of the
respective transcription start site were generated from clone
patp9SK630 (Fig. 6A). The DNA templates were linearized
with KpnI and tested for their ability to direct transcription
initiation. A 355-nt RNA species is expected upon correct initiation at
the atp9 promoter on this template (Fig. 1A).
Transcription product analysis shows that deletion of sequences
upstream of nucleotide -25 has no significant effect on the
initiation process in vitro (Fig. 6B, patp9 -67 to patp9 -25).
Deletion of additional sequences reduces specific transcription
initiation almost completely (Fig. 6B, patp9
-7 and patp9 +2). This indicates that
sequences required for transcription initiation in vitro are
clustered immediately upstream of the transcription start site within
the 25 nucleotides that are directly upstream. Although these essential
sequences contain the conserved nonanucleotide, the conserved sequence
element alone is not sufficient to efficiently promote transcription,
because deletion of the sequence between the nonanucleotide and
nucleotide -25 reduces the rate of in vitro transcription almost completely (Fig. 6B, patp9 -7).
patp9 -7 that are too weak to show up in
print.
In Vitro Transcription System for Mitochondria of Dicot
Plants
In vitro transcription systems have been
successfully established for mitochondria of a number of different
organisms including the two monocot plant species wheat and maize
(Edwards et al., 1982; Walberg and Clayton, 1983; Bogenhagen
and Yoza, 1986; Kennell and Lambowitz, 1989, Hanic-Joyce and Gray,
1991; Rapp and Stern, 1992). One of the major difficulties during the
development of an in vitro transcription system for dicot
plants was encountering the identification of a plant species amenable
as a source for the isolation of large amounts of transcriptionally
active mitochondria. We have now established such a system for pea
mitochondria that allows the investigation of various features of the
transcription initiation process in dicot plants.Accuracy of the Pea Mitochondrial Transcription
System
Every in vitro system should clearly and
accurately represent the in vivo process investigated. In this
respect the precise initiation in an in vitro transcription
system should be identical with the transcription start site determined in vivo.Competence of the Pea in Vitro Transcription System for
Promoters of Various Dicot Plants
Beyond the competent
transcription of homologous templates and templates from another
legume, soybean, the pea system initiates correctly also at promoters
from the more distantly related dicot Oenothera (Onagraceae). This includes, in addition to the
initiation sites of cox2 and atp1, the promoter for a
tRNA gene. The correct recognition of these promoters now
provides experimental evidence for a more general conservation of
promoter structures for at least two types of RNAs in mitochondria of
dicot plant species. The dissemination of the conserved nonanucleotide
motif at 5` ends of transcripts in other dicot species extends this
broad significance, although evidence for genuine promoters in these
species is still lacking (Moon et al., 1985; Young et
al., 1986; Rothenberg and Hanson, 1987). However, in dicot plant
mitochondria, transcription seems to be somehow additionally
discriminatory, because the Oenothera 18 S rRNA promoter,
although identical with respect to the conserved nonanucleotide, is not
recognized by the pea in vitro system (data not shown).
The Conserved Nonanucleotide Motif Is a Core Element of
the Dicot Plant Mitochondrial Promoters
Almost all DNA templates
carrying the conserved nonanucleotide motif
(5`-CRTAAGAGA
-3`) were
recognized in vitro even on heterologous template DNA. Because
this motif constitutes the only sequence element conserved in the
active promoter regions, the in vitro transcription analysis
strongly supports the conserved nonanucleotide motif to represent the
core element of a mitochondrial promoter structure in various dicot
plant species. This is confirmed by the deletion studies, in which only
deletion of sequences immediately upstream of the transcription start
point (i.e. 25 nucleotides preceding the transcription start
site) significantly reduces initiation activity. Although deletion of
the conserved nonanucleotide abolishes initiation completely (Fig. 6B,
patp9 +2), a small amount
of transcriptional activity remains in a construct with sequences up to
position -7 (Fig. 6B, patp9
-7, detectable on the original x-ray film). The strongly
reduced activity suggests an extended promoter structure, where the
conserved nonanucleotide motif functions as a core element
indispensable for transcription initiation and additional upstream
sequences amplify initiation efficiency.
Do Other Types of Promoters Exist in Mitochondria of
Dicot Plants?
In addition to the conserved nonanucleotide
promoters, several transcription initiation sites have been described
that show almost no similarity to this motif or to each other. An
example is the 26 S rRNA gene, whose transcription remains
controversial in dicot plant mitochondria (Binder et al.,
1994). Although transcription of this gene in potato is initiated at
the mature 5` end of the rRNA, the transcription initiation site of the Oenothera 26 S rRNA gene remains unclear. A test of both
potato and Oenothera 26 S rRNA templates in the pea in
vitro system failed to show any significant specific transcription
activity. In the pea in vitro system, only promoters
containing the conserved nonanucleotide appear to be transcribed. The
transcription start site for RNA e in soybean does not contain
the nonanucleotide motif and is not recognized in vitro.
Although located on the same template as the promoter for RNA b/c (Fig. 3A, KM2F2EH), only
RNAs with sizes that indicate initiation at the RNA b/c promoters were detected in these assays (Fig. 3B).
Additional slightly smaller RNAs observed with the KpnI-linearized template (Fig. 3B, lane
1) are most likely degradation products, because these shorter
transcripts are not present in the reaction with the HindIII-cleaved template. This observation suggests that
mitochondria of dicot plants exploit different modes of transcription
initiation using alternative promoter structures.
)
We are very grateful to Atsushi Morikami and Gregory
Brown for providing pea and soybean plasmid clones, respectively, and
to Martina Gutschow for her skilled technical assistance. We also thank
David Stern and Bill Rapp for very helpful hints during the development
of the transcription system.
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  A.-V. Bohne, S. Ruf, T. Borner, and R. Bock Faithful transcription initiation from a mitochondrial promoter in transgenic plastids Nucleic Acids Res., December 18, 2007; 35(21): 7256 - 7266. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Forner, B. Weber, S. Thuss, S. Wildum, and S. Binder Mapping of mitochondrial mRNA termini in Arabidopsis thaliana: t-elements contribute to 5' and 3' end formation Nucleic Acids Res., June 28, 2007; 35(11): 3676 - 3692. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Kuhn, A.-V. Bohne, K. Liere, A. Weihe, and T. Borner Arabidopsis Phage-Type RNA Polymerases: Accurate in Vitro Transcription of Organellar Genes PLANT CELL, March 1, 2007; 19(3): 959 - 971. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Kühn, A. Weihe, and T. Börner Multiple promoters are a common feature of mitochondrial genes in Arabidopsis Nucleic Acids Res., January 13, 2005; 33(1): 337 - 346. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Takenaka and A. Brennicke In Vitro RNA Editing in Pea Mitochondria Requires NTP or dNTP, Suggesting Involvement of an RNA Helicase J. Biol. Chem., November 28, 2003; 278(48): 47526 - 47533. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Kuhn and S. Binder RT-PCR analysis of 5' to 3'-end-ligated mRNAs identifies the extremities of cox2 transcripts in pea mitochondria Nucleic Acids Res., January 15, 2002; 30(2): 439 - 446. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J.-C. Farre and A. Araya Gene expression in isolated plant mitochondria: high fidelity of transcription, splicing and editing of a transgene product in electroporated organelles Nucleic Acids Res., June 15, 2001; 29(12): 2484 - 2491. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. M. Ikeda and M. W. Gray Characterization of a DNA-Binding Protein Implicated in Transcription in Wheat Mitochondria Mol. Cell. Biol., December 1, 1999; 19(12): 8113 - 8122. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C.-C. Chang, J. Sheen, M. Bligny, Y. Niwa, S. Lerbs-Mache, and D. B. Stern Functional Analysis of Two Maize cDNAs Encoding T7-like RNA Polymerases PLANT CELL, May 1, 1999; 11(5): 911 - 926. [Abstract] [Full Text]
|
|
|
|
|
|
S. Dombrowski, M. Hoffmann, C. Guha, and S. Binder Continuous Primary Sequence Requirements in the 18-Nucleotide Promoter of Dicot Plant Mitochondria J. Biol. Chem., April 9, 1999; 274(15): 10094 - 10099. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Mackenzie and L. McIntosh Higher Plant Mitochondria PLANT CELL, April 1, 1999; 11(4): 571 - 586. [Full Text]
|
|
|
|
|
|
D. S. Lupold, A. G. F. S. Caoile, and D. B. Stern The Maize Mitochondrial cox2 Gene Has Five Promoters in Two Genomic Regions, Including a Complex Promoter Consisting of Seven Overlapping Units J. Biol. Chem., February 5, 1999; 274(6): 3897 - 3903. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Hatzack, S. Dombrowski, A. Brennicke, and S. Binder Characterization of DNA-Binding Proteins from Pea Mitochondria Plant Physiology, February 1, 1998; 116(2): 519 - 527. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| All ASBMB Journals | Molecular and Cellular Proteomics |
| Journal of Lipid Research | ASBMB Today |