Plant Physiol. (1998) 118: 1079-1088
Circadian-Regulated Transcription of the
psbD
Light-Responsive Promoter in Wheat
Chloroplasts1
Yoichi Nakahira,
Kyoko Baba2,
Akito Yoneda,
Takashi Shiina, and
Yoshinori Toyoshima*
Graduate School of Human and Environmental Studies, Kyoto
University, Yoshida-nihonmatu-cho, Sakyo-ku, Kyoto 606-8501, Japan
 |
ABSTRACT |
The
level of mRNAs derived from the plastid-encoded psbD
light-responsive promoter (LRP) is controlled by a circadian clock(s) in wheat (Triticum aestivum). The circadian oscillations
in the psbD LRP mRNA level persisted for at least three
cycles in continuous light and for one cycle in continuous dark, with
maxima in subjective morning and minima in subjective early night. In
vitro transcription in chloroplast extracts revealed that the circadian
cycles in the psbD LRP mRNA level were dominantly
attributed to the circadian-regulated transcription of the
psbD LRP. The effects of various mutations introduced
into the promoter region on the psbD LRP activity in vitro suggest the existence of two positive elements located between 
54 and 36, which generally enhance the transcription activity, and
an anomalous core promoter structure lacking the functional "35"
element, which plays a crucial role in the circadian fluctuation and
light dependency of psbD LRP transcription activity.
| |
INTRODUCTION |
Endogenous oscillators (circadian clocks) control diurnal rhythmic
phenomena in prokaryotic cyanobacteria and almost all eukaryotes (Sweeney, 1987
; Kondo et al., 1993; Hall, 1995; Dunlap, 1996). In
plants several biological processes, including enzyme activities, leaf
movements, stomatal opening and closing, and the expression of a large
number of genes, exhibit endogenous rhythms (Sweeney, 1987; Edmunds,
1988; McClung and Kay, 1994). Photosynthesis is also one of the
representative phenomena regulated by circadian clocks at the metabolic
(Sweeney and Haxo, 1961; Hennessey and Field, 1992) and gene-expression
levels (Kloppstech, 1985; Giuliano et al., 1988; Nagy et al., 1988).
In the unicellular alga Chlamydomonas reinhardtii, several
photosynthesis-related genes encoded by both nuclear and plastidial genomes have been reported to exhibit circadian expression (Salvador et
al., 1993; Jacobshagen and Johnson, 1994; Hwang et al., 1996). However,
in higher plants all of the genes that have been shown to exhibit
circadian expression are nuclear encoded (McClung and Kay, 1994). The
possibility of circadian control over the plastid-encoded genes was
previously evaluated in pea, but no evidence for the existence of
rhythmic gene expression in the chloroplast was observed (Adamska et
al., 1991). In tomato minor diurnal fluctuations of several
plastid-encoded transcripts were detected but such fluctuations have
not been clearly defined as circadian rhythms (Piechulla, 1993).
Plastid-encoded psbD and psbC genes, encoding the
PSII subunit proteins D2 and CP43, respectively, form an operon
(psbD/C) together with other photosynthesis-related genes in
higher plants. In the wheat (Triticum aestivum)
psbD/C operon, there are four distinct transcription
initiation sites. Transcript levels derived from the third promoter
(psbD LRP) are specifically induced in etiolated seedlings
transferred to light (Wada et al., 1994). In mature wheat chloroplasts,
the accumulation of the mRNAs from the psbD LRP is
reversibly increased by illumination and reduced by dark adaptation
(Satoh et al., 1997). It has been demonstrated in wheat and other
higher plants that the light-dependent accumulation of the
psbD LRP mRNA is mediated by the light-responsive
activation/inactivation of the psbD LRP (Christopher et al.,
1992; Allison and Maliga, 1995; Hoffer and Christopher, 1997; Satoh et
al., 1997).
The level of psbD LRP mRNA may be controlled not only by
light but also by endogenous circadian rhythms. In rice seedlings some
of the mRNA levels transcribed within the psbD/C operon have been reported to exhibit diurnal oscillations under a 12-h dark/12-h light cycle (Chen et al., 1995). When etiolated wheat seedlings were
transferred to continuous light, the levels of some transcripts, which
appeared to be derived from the psbD LRP, showed a
fluctuation with an approximately 24-h periodicity for one cycle
(Kawaguchi et al., 1992). However, these data are insufficient to
confirm the involvement of circadian regulation over the
psbD LRP mRNA level. To clearly identify a circadian rhythm,
it is critical to demonstrate that a diurnal rhythm can persist for
several cycles under constant environmental conditions known as
"free-running conditions."
We examined the time course of the psbD LRP mRNA levels in
wheat seedlings under continuous light or continuous dark conditions after 5 d of 12-h light/12-h dark entrainment. The circadian
oscillations in the psbD LRP mRNA persisted for at least
3 d in continuous light and for at least 1 d in continuous
dark. Judging from the rhythms under free-running conditions, we
concluded that the psbD LRP mRNA level is controlled by a
circadian clock(s). In vitro transcription in chloroplast extracts
prepared from seedlings at several times revealed that the oscillation
in mRNA level is dominantly attributed to the circadian fluctuations of
the psbD LRP transcription activity. Furthermore, to
identify regulatory sequences involved in the clock-regulated
transcription of the psbD LRP, we investigated the effects
of various mutations introduced into the promoter region on the
psbD LRP activity in vitro.
| |
MATERIALS AND METHODS |
Plant and Growth Conditions
Wheat (Triticum aestivum) seeds were planted in moist
vermiculite and grown in a growth chamber at 25°C under the 12-h
white light (approximately 5000 lux)/12-h dark cycles. After 5 or
5.5 d, wheat seedlings were transferred to continuous light or
continuous dark conditions. Seedlings were harvested at the times
indicated in the figure legends and subjected to total cellular RNA
isolation or preparation of chloroplast extracts.
To investigate the effect of light on transcription activity in vitro,
5-d-old light-grown seedlings were maintained under dark conditions for
24 h and again transferred into light for 2 h, at which time
the chloroplast extracts were isolated.
RNA Isolation and Analysis
Primary leaves 5 to 7 cm from the top of seedlings were harvested
and quickly frozen in liquid N2. The frozen leaf
tissue was ground to a fine powder in the presence of liquid
N2. The resulting powder suspended into a buffer
(100 mM Tris-HCl, pH 8.0, 100 mM LiCl, 10 mM EDTA-NaOH, pH 8.0, and 1% SDS) was extracted twice with
an equivalent volume of a phenol mixture (phenol:chloroform:isoamyl alcohol, 25:24:1 [v/v]), and then total cellular RNA was precipitated with one-third volume of 10 M LiCl. The sample was
subjected to two additional cycles of phenol extraction followed by
ethanol precipitation.
To analyze the accumulation of plastid-encoded transcripts, S1
nuclease-protection analysis was performed as previously described (Satoh et al., 1997). The DNA probes used for detecting the
psbA and psbD LRP transcripts were a 472-bp
EcoRI-SacI fragment from pW18ES472A and a 834-bp
StyI-EcoRI fragment from pW18E866, respectively. Equal amounts of total cellular RNA (1-5 µg) were hybridized with the desired 5
-end-labeled denatured DNA probe in an 80% formamide hybridization buffer at 37°C. The RNA-DNA hybrids were treated with
400 units/mL S1 nuclease for 60 min at 20°C, and the protected DNA
fragments were separated on a denaturing polyacrylamide gel. Quantification of mRNA abundance was performed with a bio-imaging analyzer (BAS2000, Fuji, Tokyo, Japan).
Chloroplast Extract Preparation
Harvested (nonfrozen) seedlings were immediately lysed, and intact
chloroplasts were isolated by centrifugation of the cell lysate on a
10% to 80% Percoll gradient. Transcriptionally active chloroplast
extracts (high-salt extracts) were prepared from the intact
chloroplasts, as previously described (Satoh et al., 1997), and
subjected to in vitro transcription and a DNA mobility-shift assay.
Plasmid Constructs
Plasmid psbD/119 (previously designated as
pW73BE434 by Satoh et al. [1997]), containing a sequence of the
psbD LRP (119 to +315 of the transcription initiation
site) in pSP73 vector, was utilized for preparation of the DNA
constructs used in the promoter analysis of the psbD LRP. A
5 deletion fragment, which has a 5 end at 36 and a 3 end at +315
of the psbD LRP transcription initiation site, was amplified
by the PCR method with psbD/119 as a template and cloned
between the BamHI and EcoRI sites of pSP73 vector
(psbD/36). The sequence between 72 and 7 of the psbD LRP initiation site was divided into eight blocks, and
mutations were introduced in each block by means of site-directed
mutagenesis, as described by Kunkel et al. (1987). Each of the obtained
fragments were amplified by PCR and cloned between BamHI and
EcoRI sites of pSP73 (psbD/MT1 to 8). Another
mutant of the psbD LRP (psbD/MT9) having the
consensus "35" motif and the optimal spacing distance (17 bp)
between 35 and 10 was constructed by PCR. Plasmid
psbA/38, containing the sequence 38 through +278 of the
psbA transcription initiation site in the pSP72 vector was
used as a template for the psbA transcription in vitro. The
psbA promoter with a mutation in the 35 element
(psbA/MT2) was constructed by PCR from
psbA/38. For in vitro transcription, all plasmid
constructs were prepared as supercoiled circular forms.
In Vitro Transcription
In vitro transcription was carried out as previously described
(Satoh et al., 1997). Each chloroplast extract was prepared from
1.6 × 108 chloroplasts, and 1.5 µg of a given
supercoiled DNA template was incubated in a 40-µL reaction mixture at
30°C for 60 min. In vitro-synthesized RNA was isolated and analyzed
by S1 nuclease-protection analysis. A 656-bp
BamHI-NdeI fragment from pW73BE549 containing the
psbD LRP sequence (234 to +315 of the psbD LRP
transcription initiation site), followed by a 107-bp vector sequence
and a 521-bp EcoRI-HindIII fragment from
pW72ES472A containing the psbA promoter sequence (194 to
+278 of the psbA transcription initiation site), followed by
a 49-bp vector sequence were 32P-end-labeled at the
NdeI and HindIII sites of the vector sequences, respectively. These fragments were used as S1 probes for transcripts derived from the psbD LRP and psbA promoter in
vitro.
DNA Mobility-Shift Assay
For the DNA mobility-shift assay, a 32P-end-labeled,
23-bp, double-stranded oligonucleotide containing the sequence from
56 to 34 of the psbD LRP site was used as a probe.
The DNA-binding reaction was carried out in 20 µL of DEAE buffer
(Wada et al., 1994) containing 10 µL of chloroplast extract (1.0 × 108 chloroplasts per reaction), the labeled probe (0.1 ng, 1.0 × 104 cpm per reaction), and 2 µg of
poly(dI-dC)·(dI-dC) at 30°C for 30 min. The reaction mixture was
run on 5% nondenaturing polyacrylamide gel (mono:bis = 30:0.8) at
room temperature in 0.5× TBE buffer (45 mM Tris-borate, 1 mM EDTA) and then autoradiographically analyzed. The
nonlabeled DNA fragment of the probe sequence, a 26-bp fragment containing the further upstream sequence of the psbD LRP
(85 to 60 of the psbD LRP transcription initiation
site), and a 90-bp fragment of the psbD-coding sequence
were used for competition assays.
| |
RESULTS |
The Level of mRNAs Derived from the Plastid-Encoded
psbD LRP Is Controlled by a Circadian Clock(s) in Wheat
We investigated whether the fluctuation of the level of mRNAs
transcribed from the psbD LRP initiation site
(psbD LRP mRNA level) exhibits clear patterns of circadian
rhythm under free-running conditions. Wheat seedlings were grown for
5 d under a 12-h light/12-h dark regime and then shifted into
continuous light or continuous dark conditions. Seedlings were
harvested every 3 h and frozen in liquid N2.
An equivalent amount of total cellular RNA isolated from each sample
was subjected to S1 nuclease-protection analysis to determine the
relative amount of the psbD LRP mRNA. Figure 1 shows the fluctuation in the
psbD LRP mRNA level under continuous light conditions.
Sampling times are expressed as ZT (Zerr et al., 1990); ZT 0 is dawn in
the sixth cycle from imbibition. The psbD mRNA level
fluctuated with a periodicity of approximately 24 h. Starting from
ZT 0, the psbD LRP mRNA level peaked in the first subjective
morning (ZT 3), declined to a minimum in subjective early night (ZT
15), and again peaked at ZT 27 in continuous light conditions. The
psbD LRP mRNA levels varied by approximately 3- to 5-fold
between maximum and minimum. This rhythm persisted for at least three
cycles under continuous light conditions.
View larger version (27K):
[in this window]
[in a new window]
| Figure 1.
S1 nuclease-protection analyses of the
plastid-encoded psbD LRP and psbA mRNA in
wheat seedlings under continuous light. Wheat seedlings were grown for
5 d under 12-h light/12-h dark conditions and transferred to
continuous light conditions. A, S1 nuclease-protection analyses were
carried out with equal amounts of total cellular RNA (1 and 5 µg for
psbA and psbD LRP analyses, respectively)
prepared from seedlings harvested every 3 h in continuous light.
Black bars, dark periods; white bars, light periods. The inserted dark
bars indicate subjective night. S1 analysis of psbA mRNA
was not carried out for RNA samples at ZT 51 through 69. B,
Quantitative representation of the results shown in A. For each mRNA
the lowest signal is defined as 1.0 (arbitrary units), and other
abundances are expressed relative to the minimum.
|
|
The circadian cycle in the psbD LRP mRNA level was also
detected in continuous dark conditions (Fig.
2). The psbD LRP mRNA accumulation was low in the first subjective early night (approximately ZT 15) and then increased, with a peak in subjective morning (ZT 27),
and returned to a minimum in the early night phase (from ZT 39-42)
with a 3-fold amplitude. From ZT 42 onward, the accumulation of the
psbD LRP mRNA was severely reduced to the basal level, although a small peak occurred at ZT 51. Such profiles as the psbD LRP mRNA oscillation rapidly damped in continuous dark
have also been observed in the circadian expression of nuclear-encoded cab genes (Millar and Kay, 1991). That the circadian
oscillations in the psbD LRP mRNA level were detected
under continuous light and partially under continuous dark conditions
indicates that the plastid-encoded psbD LRP mRNA level is
controlled by an endogenous clock(s) in wheat.
View larger version (32K):
[in this window]
[in a new window]
| Figure 2.
S1 nuclease-protection analysis of the
plastid-encoded psbD LRP and psbA mRNA in
wheat seedlings under continuous dark conditions. Wheat seedlings were
grown for 5 d under 12-h light/12-h dark conditions, followed by
an additional 12-h light period, and were then transferred to
continuous dark conditions. A, Sampling of total cellular RNA and S1
nuclease-protection analyses were performed as described in Figure 1.
The inserted white bars indicate subjective day. ZT 0 is the initiation
of the last light period so that it is the same time as ZT 0 indicated
in Figure 1. B, Quantitative representation of the results shown in A. Relative mRNA abundances are as defined in Figure 1.
|
|
We also carried out S1 nuclease-protection analyses for the
plastid-encoded psbA transcripts encoding the D1 subunit
protein of PSII against the same RNA samples (Figs. 1 and 2). We
observed that the psbA mRNA levels appeared to fluctuate
diurnally. However, these fluctuations were very weak in both
continuous light and continuous dark conditions compared with the
psbD LRP mRNA levels.
Circadian Oscillation in the psbD LRP mRNA Level Is
Dominantly Attributed to the Clock-Regulated psbD LRP
Transcription Activity
The circadian cycle of the psbD LRP mRNA level can
result from changes at the level of transcription, mRNA stability, or
both. It has been reported that the psbD LRP mRNA levels
are regulated by light through the light-responsive
activation/inactivation of the psbD LRP (Christopher et
al., 1992; Allison and Maliga, 1995; Hoffer and Christopher, 1997;
Satoh et al., 1997). To examine the properties of circadian regulation
of the psbD LRP transcription activity, we carried out a
series of in vitro transcriptions in chloroplast extracts. Chloroplast
extracts were prepared from wheat seedlings in subjective morning
(ZT 3, 26, 27, and 51) and subjective early night (ZT 14, 15, and
39) under continuous light conditions (Fig.
3A). In vitro transcription was performed
in the extracts prepared from an equal number of chloroplasts with the
supercoiled plasmid template containing the sequences between 119 and
+315 of the psbD LRP transcription initiation site
(psbD/119). Synthesized transcripts from the
psbD LRP initiation site in vitro were specifically
detected by S1 nuclease-protection analysis. The in vitro transcription
activity of the psbD LRP fluctuated diurnally throughout
experimental periods, with about a 10-fold increase in subjective
morning (ZT 3, 26, 27, and 51) compared with that in subjective early
night (ZT 14, 15, and 39). Although we did not determine the time
course of the circadian fluctuation in psbD LRP
transcription in detail, this result indicates that the transcription
activity of the psbD LRP is under circadian control and
that the circadian oscillation in the psbD LRP mRNA level is dominantly attributed to the clock-regulated
psbD LRP transcription activity.
View larger version (41K):
[in this window]
[in a new window]
| Figure 3.
In vitro transcription of the psbD
LRP and the psbA promoter in chloroplast extracts
prepared from wheat seedlings in continuous light conditions. A, In
vitro transcription from the psbD LRP transcription
initiation site was carried out in two series with equal amounts of the
chloroplast high-salt extracts prepared from seedlings at the indicated
times (ZT 3, 14, 15, 26, 27, and ZT 3, 15, 27, 39, and 51, respectively) with the supercoiled circular plasmid pSP73 vector
containing the sequences between 119 and +315 of the
psbD LRP transcription initiation site
(psbD/119) as the template. Specific in vitro
transcripts were detected by S1 nuclease-protection analysis. B, In
vitro transcription of the psbA promoter was carried out
with one of two series of the chloroplast extracts (ZT 3, 14, 15, 26, and 27) and circular plasmid pSP72 vector containing the sequences
between 38 and +278 of the psbA transcription
initiation site (psbA/38) as the template. In vitro
transcripts were subjected to S1 nuclease-protection analyses. The
arrows indicate the primary transcripts synthesized in vitro.
Read-through signals (*) were also seen at the position of the boundary
between the vector and the inserted plastid DNA sequences.
|
|
We also assayed in vitro transcription activity of the
psbA gene with a plasmid template containing the
sequence between 38 and +278 of the psbA transcription
initiation site (psbA/38). In vitro transcription
of the psbA promoter with the same series of
chloroplast extracts showed the diurnal fluctuation of the psbA transcription activity with 2- to 3-fold increases
in subjective morning compared with that in subjective early night
(Fig. 3B). This result suggests that, in addition to the
psbD LRP, the psbA transcription may also
be controlled by an endogenous clock(s).
Two cis Elements at the 5 Upstream Region in
the psbD LRP May Act as Positive Elements That Generally
Enhance the Transcription Activity Rather Than Circadian
Clock-Responsive Elements
The psbD LRP contains a 60-bp region upstream of
the transcription initiation site that is highly conserved among
several monocotyledonous and dicotyledonous plants (Fig.
4). This region possesses two positive
elements involved in the light-dependent transcription of the
psbD LRP in barley and tobacco. One is an AAGT repeat
located between 54 and 46 of the transcription initiation site (Kim
and Mullet, 1995). Another is a GACC/T repeat located between 44 and
36 (Allison and Maliga, 1995). We carried out in vitro transcription
in wheat chloroplast extracts from 24-h dark-adapted and 2-h
reilluminated seedlings with a series of DNA constructs containing
mutations in the upstream region of the psbD LRP
initiation site. The results indicated that the same cis
elements function as positive regulators in the light-induced transcription of the wheat psbD LRP (data not shown).
View larger version (21K):
[in this window]
[in a new window]
| Figure 4.
Highly conserved sequences between the wheat
psbD LRP region and homologous regions from other higher
plants. DNA sequences between 60 and +20 of the wheat
psbD LRP transcription initiation site are aligned with
barley, rice, tobacco, and Arabidopsis (Christopher et al., 1992;
Hoffer and Christopher, 1997). Completely conserved sequences are
boxed. The positive regulatory elements, the AAGT and GACC/T repeats,
which have been indicated to be involved in light-dependent
transcription of the psbD LRP (Allison and Maliga, 1995;
Kim and Mullet, 1995), and the putative 35 and 10 elements are
marked.
|
|
To investigate whether the cis elements would also be
involved in circadian regulation of the psbD LRP, we
examined the effect of mutations introduced into the two elements on
the in vitro transcription activity in chloroplast extracts prepared
from seedlings under continuous light conditions. Introduction of
mutations in each element (psbD/MT3 and 4) resulted
in a 2-fold reduction of transcription activity, but the transcription
levels fluctuated markedly (Fig. 5B).
Complete deletion of both elements (psbD/36) resulted
in a severe reduction of the in vitro transcription activity in
chloroplast extracts prepared from seedlings in subjective morning
(Fig. 5C). However, it should be noted that the
psbD/36 construct still conferred significant
circadian fluctuations in the transcription levels initiated from the
psbD LRP site.
View larger version (44K):
[in this window]
[in a new window]
| Figure 5.
Effects on the circadian fluctuation of the in
vitro psbD LRP transcription activity with the mutations
and deletions of the positive regulatory elements. A, DNA fragments
(119 to +315) containing the indicated base-pair substitutions
(psbD/MT3 and 4) and a DNA fragment deletion
containing sequences between 36 and +315 (psbD/36)
were cloned in pSP73 vector and used as templates. B and C, In vitro
transcription was carried out in chloroplast extracts prepared from
seedlings at indicated times in continuous light conditions. Details
are as described in Figure 3.
|
|
A putative transcription factor that specifically recognizes the AAGT
repeat has been suggested in barley chloroplasts (Kim and Mullet,
1995). A DNA-binding protein that recognizes the GACC/T repeat in
tobacco chloroplasts has been reported (Allison and Maliga, 1995). To
reveal the role of such trans factor(s) on the circadian
regulation and the light-dependent behavior of the psbD LRP, we performed a DNA mobility-shift assay using a 23-bp DNA fragment
probe containing the sequence from 56 through 34 of the
psbD LRP initiation site. Incubation of chloroplast
extract prepared from seedlings at ZT 3 in continuous light with the
DNA probe resulted in the formation of several DNA/protein complexes (Fig. 6A). Among them, bands "a" and
"d" were apparently due to interactions with sequence-specific
DNA-binding protein(s), since the nonlabeled DNA fragment of the probe
sequence could specifically act as a competitor. However, the relative
intensities of the a and d bands revealed neither circadian
fluctuations (Fig. 6B) nor light dependency (data not shown). Together
with the results shown in Figure 5, the sequence between 54 and 36
is suggested to function as two positive elements that generally
enhance the psbD LRP transcription activity rather than
acting as essential elements for circadian regulation.
View larger version (40K):
[in this window]
[in a new window]
| Figure 6.
DNA mobility-shift assay of the upstream positive
elements with chloroplast extracts prepared from seedlings in
continuous light conditions. A, End-labeled, 23-bp, double-stranded DNA
probe (56 to 34) was incubated in the absence (lane 1) and in the
presence (lanes 2-11) of chloroplast extract prepared from seedlings
at ZT 3 in continuous light. Binding reactions were performed in the
absence (lanes 1 and 2) or in the presence at the indicated times of
molecular excess of the unlabeled probe sequence (lanes 35), a 26-bp
DNA fragment containing another region of psbD LRP (85
to 60) not involved in transcription activity (lanes 68), or a
90-bp psbD-coding region sequence (lanes 911) as DNA
competitors. Arrows mark detected complexes. B, Binding reactions
carried out in chloroplast extracts prepared from seedlings at the
indicated times in continuous light conditions (ZT 3, 15, 27, 39, and
51).
|
|
The Core Promoter Structure Lacking the Prokaryotic 35 Element
May Play a Crucial Role in the Circadian Fluctuation and Light
Dependency of the psbD LRP Transcription Activity
Typical plastid promoters contain sequences resembling prokaryotic
10 (TATAAT) and 35 (TTGACA) elements that are structurally and
functionally similar to Escherichia coli 
70-type
promoters (Hanley-Bowdoin and Chua, 1987). The importance of the
E. coli-like elements for transcription initiation of
plastid genes has been demonstrated by in vitro mutational analyses
(Link, 1984; Bradley and Gatenby, 1985; Gruissem and Zurawski, 1985).
The psbD LRP also contains similar elements (Fig. 4), but
the putative 35 element (TTTAAT) only weakly resembles the
prokaryotic consensus motif, and the spacing between the 35 and 10
elements is anomalous (15 bp instead of the optimal 17-19 bp).
To investigate the relationship between the psbD LRP core
promoter sequence and the circadian fluctuation of the psbD
LRP transcription, we constructed several DNA constructs containing mutations in the core promoter region of the psbD LRP.
Complete destruction of the putative 35 element (32 to 28 of the
psbD LRP transcription initiation site,
psbD/MT5; Fig. 7B) or the spacing sequences between the 35 and 10 elements (27 to 13, psbD/MT6 and 7; data not shown) caused no change in the
psbD LRP transcription activity in continuous light
conditions. Conversely, destruction of the 10 element (12 to 7,
psbD/MT8) resulted in an almost disappearance of the
psbD LRP activity (Fig. 7B). Furthermore, we carried out in
vitro transcription with the same set of DNA constructs in chloroplast
extracts from 24-h dark-adapted and 2-h reilluminated wheat seedlings
("D" and "L," respectively), and the results were similar to
those for circadian regulation (Fig. 7C). These results indicate that
the 10 element is essential for the basal transcription activity, but
neither the sequence corresponding to the putative 35 element nor the
spacing sequence between two elements plays any positive role in
psbD LRP transcription behavior.
View larger version (40K):
[in this window]
[in a new window]
| Figure 7.
Effects of the generated mutations in the core
promoter region of the psbD LRP on its circadian- and
light-regulated in vitro transcription. A, DNA fragments (119 to
+315) containing the indicated base-pair substitutions at each block
were cloned in pSP73 vector (psbD/MT5 to 8) and used
as templates. B, In vitro transcription was carried out in chloroplast
extracts prepared from seedlings at the indicated times in continuous
light conditions (ZT 3, 15, and 27). Results for templates
psbD/MT6 and 7 are not shown. C, In vitro
transcription with the same set of templates was carried out in
chloroplast extracts prepared from 24-h dark-adapted and then 2-h
reilluminated seedlings (D and L extracts, respectively). Read-through
signals (*) were also seen at the position of the boundary between the
vector and the inserted plastid DNA sequences.
|
|
However, it might be possible that the anomalous structure of the
psbD LRP lacking the functional 35 elements is responsible for the circadian-fluctuation dependency and/or the light dependency of
the psbD LRP transcription. To examine this possibility we changed the core promoter structure of the psbD LRP to
resemble a typical prokaryotic promoter by substituting the pseudo 35 elements with consensus 35 motif and simultaneously converting the
spacing distance between 35 and 10 to 17 bp
(psbD/MT9). This construct abolished the circadian
fluctuation of in vitro transcription activity from the psbD
LRP initiation site, leaving the promoter to function almost
constitutively in continuous light (Fig.
8B). A similar profile was observed in
the light dependency of the psbD LRP transcription activity
(Fig. 8C). These results suggest that the promoter architecture lacking
the 35 element plays a crucial role in the circadian fluctuation and
light dependency of the psbD LRP transcription activity.
View larger version (41K):
[in this window]
[in a new window]
| Figure 8.
Effects on the in vitro psbD LRP
transcription activity after modification of the psbD
LRP to generating a consensus 35 element and canonical spacing (17 bp). A, DNA fragment (119 to +315) containing the indicated base-pair
substitutions and insertion was cloned in pSP73 vector
(psbD/MT9) and used as templates. B, In vitro
transcription carried out in chloroplast extracts prepared from
seedlings at the indicated times in continuous light conditions (ZT 3, 14, 15, 26, and 27). C, In vitro transcription performed in chloroplast
extracts prepared from 24-h dark-adapted (D) and 2-h reilluminated (L)
seedlings, as described in Figure 7C. Read-through signals (*) were
also seen at the position of the boundary between the vector and the
inserted plastid DNA sequences.
|
|
The psbA promoter contains typical 35 and 10 elements,
and the spacing between them is optimal (18 bp). Exchange of the 35 element (TTGACA) of the psbA promoter with the pseudo
35 element (TTTAAT) of the psbD LRP resulted in very
strong circadian fluctuation of the psbA transcription
activity (Fig. 9B). Severe reduction of
the transcription levels was observed specifically in chloroplast
extracts prepared from seedlings in subjective early night (ZT 14 and
15), with almost no effect in subjective morning (ZT 3, 26, and 27).
Furthermore, light induction of the psbA transcription
activity also became stronger after the 35 element exchange (Fig.
9C). These results suggest that the robust circadian fluctuation and
light induction of the transcription activity observed in the
psbD LRP, but not in the psbA promoter, may be
attributed to the difference of their core promoter structures, which
lack or possess the functional 35 element, respectively.
View larger version (37K):
[in this window]
[in a new window]
| Figure 9.
Effects on the in vitro psbA
transcription activity after substituting the 35 element of the
psbA promoter into the pseudo element of
psbD LRP. A, DNA fragment between 38 and +278
containing the indicated base-pair substitutions in the 35 element
was cloned in pSP72 vector and used as templates
(psbA/MT2). B, In vitro transcription was carried out
in chloroplast extracts prepared from seedlings at indicated times in
continuous light conditions (ZT 3, 14, 15, 26, and 27). C, In vitro
transcription performed in chloroplast extracts prepared from 24-h
dark-adapted (D) and 2-h reilluminated (L) seedlings, as described in
Figure 7C. Read-through signals (*) were also seen at the position of
the boundary between the vector and the inserted plastid DNA
sequences.
|
|
| |
DISCUSSION |
The psbD LRP Transcription Is under Circadian Control
in Chloroplasts of Higher Plants
In this report we have demonstrated that the levels of the
plastid-encoded psbD LRP mRNAs, which are derived from the
third transcription initiation site (psbD LRP site) in the
psbD/C operon, are conspicuously controlled by a circadian
clock(s) in wheat. The circadian oscillation in the psbD LRP
mRNA level persisted for at least three cycles in continuous light and
one cycle in continuous dark, with a maximum in subjective morning and
a minimum in subjective early night (Figs. 1 and 2). In vitro
transcription activity of the psbD LRP in chloroplast
extracts prepared from the wheat seedlings under continuous light
diurnally fluctuated throughout experimental periods, with about a
10-fold increase in subjective morning compared with that in subjective
early night (Fig. 3A). In vitro chloroplast transcription assays in
wheat reproduced the light-dependent psbD LRP transcription
pattern in vivo (Wada et al., 1994; Satoh et al., 1997). Therefore, it is likely that the results shown in Figure 3A indicate that the circadian oscillations of the psbD LRP mRNA expression is
mostly attributable to the transcriptional regulation of the
psbD LRP.
Since the promoter structure of the psbD LRP is highly
conserved among certain higher plants (Christopher et al., 1992; Hoffer and Christopher et al., 1997; Fig. 4), circadian oscillation in the
psbD LRP transcription activity is likely to be exhibited in
other higher plants as well. The psbD LRP has been shown to be activated by high-intensity blue and UV-A light in barley
chloroplasts (Gamble and Mullet, 1989; Christopher and Mullet, 1994),
and it has been proposed that the high-irradiance activation of the
psbD LRP serves to maintain D2 protein levels that are
photodamaged under high-intensity light conditions (Christopher and
Mullet, 1994). In addition to such a role, the circadian oscillation of the psbD LRP may also control the synthetic rate of D2
protein in a diurnal manner.
Circadian-regulated transcription of the plastid-encoded genes in
higher plants may not be restricted to the psbD LRP.
Although the mRNA level of the plastid-encoded psbA gene
showed only weak oscillation with less than a 2-fold variation (Figs. 1
and 2), in vitro transcription activity of psbA promoter
fluctuated diurnally with 2- to 3-fold variation (Fig. 3B). Since the
psbA mRNA is highly stable (t1/2 = 10-40 h) in mature leaves of higher plants (Mullet and Klein, 1987;
Klaff and Gruissem, 1991; Kim et al., 1993), the psbA mRNA
level may not reflect the fluctuation of psbA transcription
activity.
Activity of a RNA Polymerase, Which Does Not Require the 35
Element for the psbD LRP and psbA Promoter
Recognition, May Be Regulated by a Circadian Clock(s) and Light
In vitro transcription with the mutated psbD LRP
sequences revealed the existence of two cis elements and
a core promoter structure that are involved in the circadian regulation
of the psbD LRP activity. The two cis
elements are an AAAGTAAGT sequence located between 54 and 46 of the
wheat psbD LRP transcription initiation site and a
GACCTGACT sequence between 44 and 36. These elements are identical
to those demonstrated to function as positive regulatory elements of
the psbD LRP in tobacco and barley (Allison and Maliga,
1995; Kim and Mullet, 1995). Mutations introduced into the two elements
resulted in a reduction of the in vitro psbD LRP
activity, but significant circadian fluctuations in the
psbD LRP transcription levels remained (Fig. 5). The
ability of DNA-binding factor(s) recognizing these elements was
constitutive in continuous light (Fig. 6). These results suggest that
the two cis elements act as positive regulators that
generally enhance the psbD LRP transcription activity
rather than the circadian clock-responsive elements controlling the
oscillation.
In contrast, alterations in the core-promoter sequence of the
psbD LRP suggest that the core-promoter architecture,
lacking the functional 35 element, plays a crucial role in the
circadian fluctuation of the psbD LRP transcription
activity (Figs. 7B and 8B). A mutation in the psbA core
promoter, in which the functional 35 element of the
psbA promoter was replaced by the pseudo 35 element
from the psbD LRP, resulted in more robust fluctuation of the in vitro psbA transcription activity than the
native psbA promoter (Fig. 9B). This suggests that the
core promoter structures of the psbA promoter and the
psbD LRP, possessing the functional 35 element or not,
respectively, may be responsible for the different robustness of
fluctuations of the psbA promoter (2- to 3-fold) and
psbD LRP (10-fold). Furthermore, the same core promoter
structure of the psbD LRP, lacking the functional 35
element, is suggested to be responsible for the light-responsive
behavior of psbD LRP transcription activity. Thus, it is
likely that the circadian-clock and light-responsive element(s) in the
psbD LRP are common.
As a possible mechanism that enables the psbD LRP
lacking the functional 35 element to be under circadian and light
regulation, we speculate that the chloroplast contains at least two
types of transcription apparatus that either require the functional 35 element for promoter recognition or do not. Although activity of
the former type of RNA polymerase may not be clearly regulated by a
circadian clock(s) and light, the latter type may be under strong
circadian and light regulation. This working hypothesis can easily
explain the strong circadian and light regulation of the
psbD LRP and the mutated psbA promoter,
in which the functional 35 element of the psbA
promoter was replaced by the pseudo 35 element from the
psbD LRP. However, the native psbA
promoter, which possesses the functional 35 element, can be
recognized by both types of RNA polymerase; therefore, circadian
control and light induction of psbA transcription may
not be so clearly observed. Since it was reported that the
psbD LRP mRNA disappeared in transgenic tobacco lacking
a subunit of the plastid-encoded plastid RNA polymerase (PEP), which is
similar to RNA polymerase in eubacteria (Allison et al., 1996), our
proposed circadian and light-regulated RNA polymerase activity may be
mediated through a circadian- and/or light-dependent regulatory
mechanism(s) of the PEP.
The PEP comprises 
-, 
-, -, and "-subunits, which are
encoded by rpoA, rpoB,
rpoC1, and rpoC2, respectively, in the
plastid genome. These subunits form a core enzyme that is structurally and functionally homologous to the bacterial enzyme. In addition to its
core enzyme, a bacterial RNA polymerase contains one of the factors
that are essential for transcription initiation and promoter
selectivity (Helmann and Chamberlin, 1988). In higher plants the
chloroplast factors are encoded by nuclear genomes (Isono et al.,
1997; Tanaka et al., 1997; Kestermann et al., 1998; Tozawa et al.,
1998). Recently, we also cloned a putative chloroplast factor
(SigA) from wheat, whose mRNA level is regulated by
light (K. Morikawa, Y. Tsunoyama, T. Shiina, and Y. Toyoshima,
unpublished data). Northern analysis with a SigA probe
revealed that the SigA mRNA exhibited circadian
oscillation with phase patterns similar to that of the
psbD LRP mRNA level in continuous light (Y. Nakahira, T. Shiina, and Y. Toyoshima, unpublished data). These results led us to
propose a mechanism for the circadian and light regulation of the
plastid-encoded psbD LRP transcription with which the
psbD LRP transcription activity is controlled by the
circadian- and light-dependent expression of a nuclear-encoded factor. In fact, the rpoD2 encoding a 70-like
transcription factor in cyanobacteria has been shown to be essential
for the circadian-regulated transcription of a subset of genes
(Tsinoremas et al., 1996).
Alternatively, a unique model that accounts for light-responsive
regulation of PEP activity has been described for mustard plastids
(Tiller and Link, 1993). With this model low-phosphorylated PEP can
effectively initiate transcription in the chloroplast, whereas the
highly phosphorylated PEP forms a tightly bound complex among the PEP
core enzyme, -like factors, and the promoter sequence that arrests
transcription in the etioplast. Recently, the light-dependent regulation of phosphorylation state in the nucleus and cytoplasm was
shown to be required for the blue-light-responsive transcription of the
barely psbD LRP (Christopher et al., 1997), and it
was proposed that the light-dependent phosphorylation/dephosphorylation event(s) outside of the plastids may control the phosphorylation state
of PEP, which recognizes the psbD LRP through an
interorganelle pathway. The same phosphorylation/dephosphorylation
event(s) may also be implicated in the circadian regulation of
psbD LRP transcription.
| |
FOOTNOTES |
1
This work was supported by a Grant-in-Aid for
Scientific Research on Priority Areas (no. 10170218 to Y.T.); by
Grants-in-Aid for Scientific Research from the Ministry of Education,
Science, and Culture of Japan (no. 09874170 to Y.T. and nos. 10640629, 10309008 to T.S.); and by a grant from the Research for the Future Program, Japan Society for the Promotion of Science (no.
JSPS-RFTF96L00604 to T.S.).
2
Present address: The Institute of Physical and
Chemical Research (RIKEN), 2-1 Hirosawa, Wako 351-0106, Japan.
*
Corresponding author; e-mail
toyoshima{at}soumu1.jinkan.kyoto-u.ac.jp; fax 81-75-753-6577.
Received April 23, 1998;
accepted August 17, 1998.
| |
ABBREVIATIONS |
Abbreviations:
LRP, light-responsive promoter.
PEP, plastid-encoded plastid RNA polymerase.
ZT, Zeitgeiber time.
| |
ACKNOWLEDGMENTS |
We are grateful to Prof. Y. Isozumi (Kyoto University) for
allowing us to use the facilities of the Radioisotope Research Center.
We also thank Dr. M.V. Lemas for critical reading of the manuscript.
| |
LITERATURE CITED |
Adamska I,
Scheel B,
Kloppstech K
(1991)
Circadian oscillations of nuclear-encoded chloroplast proteins in pea (Pisum sativum).
Plant Mol Biol
17:
1055-1065
[CrossRef][ISI][Medline]
Allison LA,
Maliga P
(1995)
Light-responsive and transcription-enhancing elements regulate the plastid psbD core promoter.
EMBO J
14:
3721-3730
[ISI][Medline]
Allison LA,
Simon LD,
Maliga P
(1996)
Deletion of rpoB reveals a second distinct transcription system in plastids of higher plants.
EMBO J
15:
2802-2809
[ISI][Medline]
Bradley D,
Gatenby A
(1985)
Mutational analysis of the maize chloroplast ATPase-subunit gene promoter: the isolation of promoter mutants in E. coli and their characterization in a chloroplast in vitro transcription system.
EMBO J
4:
3641-3648
[ISI][Medline]
Chen SCG,
Wu SP,
Lo PK,
Mon DP,
Chen LFO
(1995)
Regulation of plastid photosynthetic psbK-I-D-C gene expression by light in rice plants.
Physiol Plant
93:
617-623
[CrossRef]
Christopher DA,
Kim M,
Mullet JE
(1992)
A novel light-regulated promoter is conserved in cereal and dicot chloroplasts.
Plant Cell
4:
785-798
[Abstract/Free Full Text]
Christopher DA,
Mullet JE
(1994)
Separate photosensory pathways co-regulate blue light/ultraviolet-A-activated psbD-psbC transcription and light-induced D2 and CP43 degradation in barley (Hordeum vulgare) chloroplasts.
Plant Physiol
104:
1119-1129
[Abstract]
Christopher DA,
Xinli L,
Kim M,
Mullet JE
(1997)
Involvement of protein kinase and extraplastidic serine/threonine protein phosphatases in signaling pathways regulating plastid transcription and the psbD blue light-responsive promoter in barley.
Plant Physiol
113:
1273-1282
[Abstract]
Dunlap JC
(1996)
Genetic and molecular analysis of circadian rhythms.
Annu Rev Genet
30:
579-601
[CrossRef][ISI][Medline]
Edmunds LN
(1988)
Cellular and Molecular Bases of Biological Clocks.
Springer-Verlag, New York
Gamble PE,
Mullet JE
(1989)
Blue light regulates the accumulation of two psbD-psbC transcripts in barley chloroplasts.
EMBO J
8:
2785-2794
[ISI][Medline]
Giuliano G,
Hoffman NE,
Ko K,
Scolnik PA,
Cashmore AR
(1988)
A light-entrained circadian clock controls transcription of several plant genes.
EMBO J
7:
3635-3642
[ISI][Medline]
Gruissem W,
Zurawski G
(1985)
Analysis of promoter regions for the spinach chloroplast rbcL, atpB and psbA genes.
EMBO J
4:
3375-3383
[ISI][Medline]
Hall JC
(1995)
Tripping along the trial to the molecular mechanisms of biological clocks.
Trends Neurosci
18:
230-240
[CrossRef][ISI][Medline]
Hanley-Bowdoin L,
Chua NH
(1987)
Chloroplast promoters.
Trends Biochem Sci
12:
67-70
Helmann JD,
Chamberlin MJ
(1988)
Structure and function of bacterial sigma factors.
Annu Rev Biochem
57:
839-872
[CrossRef][ISI][Medline]
Hennessey TL,
Field CB
(1992)
Evidence of multiple circadian oscillators in bean plants.
J Biol Rhythms
7:
105-113
[Abstract/Free Full Text]
Hoffer PH,
Christopher DA
(1997)
Structure and blue-light-responsive transcription of a chloroplast psbD promoter from Arabidopsis thaliana.
Plant Physiol
115:
213-222
[Abstract]
Hwang S,
Kawazoe R,
Herrin DL
(1996)
Transcription of tufA and other chloroplast-encoded genes is controlled by a circadian clock in Chlamydomonas.
Proc Natl Acad Sci USA
93:
996-1000
[Abstract/Free Full Text]
Isono K,
Shimizu M,
Yoshimoto K,
Niwa Y,
Satoh K,
Yokota A,
Kobayashi H
(1997)
Leaf-specifically expressed genes for polypeptides destined for chloroplasts with domains of sigma70 factors of bacterial RNA polymerases in Arabidopsis thaliana.
Proc Natl Acad Sci USA
94:
14948-14953
[Abstract/Free Full Text]
Jacobshagen S,
Johnson CH
(1994)
Circadian rhythms of gene expression in Chlamydomonas reinhardtii: circadian cycling of mRNA abundances of cabII, and possibly of -tubulin and cytochrome c.
Eur J Cell Biol
64:
142-152
[Medline]
Kawaguchi H,
Fukuda I,
Shiina T,
Toyoshima Y
(1992)
Dynamic behavior of psb gene transcripts in greening wheat seedlings. I. Time course of accumulation of the psbA through psbN gene transcripts during light-induced greening.
Plant Mol Biol
20:
695-704
[ISI][Medline]
Kestermann M,
Neukirchen S,
Kloppstech K,
Link G
(1998)
Sequence and expression characteristics of a nuclear-encoded chloroplast sigma factor from mustard.
Nucleic Acids Res
26:
2747-2754
[Abstract/Free Full Text]
Kim M,
Christopher DA,
Mullet JE
(1993)
Direct evidence for selective modulation of psbA, rpoA, rbcL and 16S RNA stability during barley chloroplast development.
Plant Mol Biol
22:
447-463
[CrossRef][ISI][Medline]
Kim M,
Mullet JE
(1995)
Identification of a sequence-specific DNA binding factor required for transcription of the barley chloroplast blue light-responsive psbD-psbC promoter.
Plant Cell
7:
1445-1457
[Abstract]
Klaff P,
Gruissem W
(1991)
Changes in chloroplast mRNA stability during leaf development.
Plant Cell
3:
517-529
[Abstract/Free Full Text]
Kloppstech K
(1985)
Diurnal and circadian rhythmicity in the expression of light-induced plant nuclear messenger RNAs.
Planta
165:
502-506
[CrossRef][ISI]
Kondo T,
Strayer CA,
Kulkarni RD,
Taylor W,
Ishiura M,
Golden SS,
Johnson CH
(1993)
Circadian rhythms in prokaryotes: luciferase as a reporter of circadian gene expression in cyanobacteria.
Proc Natl Acad Sci USA
90:
5672-5676
[Abstract/Free Full Text]
Kunkel TA,
Roberts JD,
Zakour RA
(1987)
Rapid and efficient site-specific mutagenesis without phenotypic selection.
Methods Enzymol
154:
367-382
[ISI][Medline]
Link G
(1984)
DNA sequence requirements for the accurate transcription of a protein-coding plastid gene in a plastid in vitro system from mustard (Sinapis alba L.).
EMBO J
3:
1697-1704
[ISI][Medline]
McClung CR,
Kay SA
(1994)
Circadian rhythms in Arabidopsis thaliana.
In
EM Meyerowitz,
CR Somerville,
eds, Arabidopsis.
Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY, pp 615-637
Millar AJ,
Kay SA
(1991)
Circadian control of cab gene transcription and mRNA accumulation in Arabidopsis.
Plant Cell
3:
541-550
[Abstract/Free Full Text]
Mullet J,
Klein R
(1987)
Transcription and mRNA stability are important determinants of higher plants chloroplast RNA levels.
EMBO J
6:
1571-1579
[ISI][Medline]
Nagy F,
Kay SA,
Chua NH
(1988)
A circadian clock regulates transcription of the wheat Cab-1 gene.
Genes Dev
2:
376-382
[Abstract/Free Full Text]
Piechulla B
(1993)
`Circadian clock' directs the expression of plant genes.
Plant Mol Biol
22:
533-542
[CrossRef][ISI][Medline]
Salvador ML,
Klein U,
Bogorad L
(1993)
Light-regulated and endogenous fluctuations of chloroplast transcript levels in Chlamydomonas. Regulation by transcription and RNA degradation.
Plant J
3:
213-219
[CrossRef][ISI][Medline]
Satoh J,
Baba K,
Nakahira Y,
Shiina T,
Toyoshima Y
(1997)
Characterization of dynamics of the psbD light-induced transcription in mature wheat chloroplasts.
Plant Mol Biol
33:
267-278
[CrossRef][ISI][Medline]
Sweeney BM
(1987)
Rhythmic Phenomena in Plants.
Academic Press, San Diego, CA
Sweeney BM,
Haxo FT
(1961)
Persistence of a photosynthetic rhythm in enucleated Acetabularia.
Science
134:
1361-1363
[Abstract/Free Full Text]
Tanaka K,
Tozawa Y,
Mochizuki N,
Shinozaki K,
Nagatani A,
Wakasa K,
Takahashi H
(1997)
Characterization of three cDNA species encoding plastid RNA polymerase sigma factors in Arabidopsis thaliana: evidence for the sigma factor heterogeneity in higher plant plastids.
FEBS Lett
413:
309-313
[CrossRef][ISI][Medline]
Tiller K,
Link G
(1993)
Phosphorylation and dephosphorylation affect functional characteristics of chloroplast and etioplast transcription systems from mustard (Sinapis alba L.).
EMBO J
12:
1745-1753
[ISI][Medline]
Tozawa Y,
Tanaka K,
Takahashi H,
Wakasa K
(1998)
Nuclear encoding of a plastid factor in rice and its tissue- and light-dependent expression.
Nucleic Acids Res
26:
415-419
[Abstract/Free Full Text]
Tsinoremas NF,
Ishiura M,
Kondo T,
Andersson CR,
Tanaka K,
Takahashi H,
Johnson CH,
Golden SS
(1996)
A sigma factor that modifies the circadian expression of a subset of genes in cyanobacteria.
EMBO J
15:
2488-2495
[ISI][Medline]
Wada T,
Tunoyama Y,
Shiina T,
Toyoshima Y
(1994)
In vitro analysis of light-induced transcription in the wheat psbD/C gene cluster using plastid extracts from dark-grown and short- term-illuminated seedlings.
Plant Physiol
104:
1259-1267
[Abstract]
Zerr DM,
Hall JC,
Rosbash M,
Siwicki KK
(1990)
Circadian fluctuations of period protein immunoreactivity in the CNS and the visual system of Drosophila.
J Neurosci
10:
2749-2762
[Abstract]
Top
Abstract
Introduction
