Anda di halaman 1dari 12

Plant Mol Biol (2013) 82:5970

DOI 10.1007/s11103-013-0037-0

Plastid gene expression during chloroplast differentiation

and dedifferentiation into non-photosynthetic plastids during
seed formation
Guillaume Allorent Florence Courtois
Fabien Chevalier Silva Lerbs-Mache

Received: 22 January 2013 / Accepted: 28 February 2013 / Published online: 15 March 2013
Springer Science+Business Media Dordrecht 2013

Abstract Arabidopsis seed formation is coupled with two

plastid differentiation processes. Chloroplast formation
starts during embryogenesis and ends with the maturation
phase. It is followed by chloroplast dedifferentiation/
degeneration that starts at the end of the maturation phase
and leads to the presence of small non-photosynthetic
plastids in dry seeds. We have analysed mRNA and protein
levels of nucleus- and plastid-encoded (NEP and PEP)
components of the plastid transcriptional machinery,
mRNA and protein levels of some plastid RNA polymerase
target genes, changes in plastid transcriptome profiles and
mRNA and protein levels of some selected nucleus-encoded plastid-related genes in developing seeds during
embryogenesis, maturation and desiccation. As expected,
most of the mRNAs and proteins increase in abundance
during maturation and decrease during desiccation, when
plastids dedifferentiate/degenerate. In contrast, mRNAs
and proteins of components of the plastid transcriptional
apparatus do not decrease or even still increase during the
period of plastid dedifferentiation. Results suggest that
proteins of the plastid transcriptional machinery are specifically protected from degradation during the desiccation
period and conserved in dry seeds to allow immediate
regain of plastid transcriptional activity during stratification/germination. In addition, results reveal accumulation
and storage of mRNAs coding for RNA polymerase

Electronic supplementary material The online version of this

article (doi:10.1007/s11103-013-0037-0) contains supplementary
material, which is available to authorized users.
G. Allorent  F. Courtois  F. Chevalier  S. Lerbs-Mache (&)
Laboratoire de Physiologie Cellulaire Vegetale, iRTSV, UMR
5168, CNRS/UJF/CEA/INRA, CEA-Grenoble, 17 rue des
Martyrs, 38054 Grenoble cedex, France

components and sigma factors in dry seeds. They should

provide immediately-to-use templates for translation on
cytoplasmic ribosomes in order to enhance RNA polymerase protein levels and to provide regulatory proteins for
stored PEP to guaranty efficient plastid genome transcription during germination.
Keywords Arabidopsis  Plastids  Dedifferentiation 
Transcriptome profiling  RNA polymerase

Plant cells contain plastids that differentiate into functionally distinct forms (e.g. amyloplasts, chromoplasts,
leucoplasts, chloroplasts ect.) depending on the cellular and
developmental context of the plant. Plastid differentiation
is accompanied by changes in morphology, structure and
plastid gene expression. The most studied plastid type is
the chloroplast that is active in photosynthesis, and plastid
gene expression is well characterized in chloroplasts. Much
less is known concerning gene expression in non-green
plastids and changes in gene expression during plastid
differentiation. Rather nothing is known concerning plastid
dedifferentiation. Actually, in-depth plastid transcriptome
profiling has been performed to monitor tomato fruit
plastid differentiation (i.e. chloroplast to chromoplast
transformation, Kahlau and Bock 2008), to characterize
potato tuber amyloplasts (Valkov et al. 2009) and to follow
the differentiation of seed eoplasts into amyloplastst and
chloroplasts during Arabidopsis germination and early
plant development (Demarsy et al. 2012).
In order to get also knowledge on plastid dedifferentiation, in the present article we have chosen Arabidopsis
seed formation as biological system. Seed formation of



higher plants begins with a double fertilization process and

ends with the establishment of a mature dry seed by
passing through three well-defined developmental stages,
i.e. embryogenesis, maturation and desiccation (for schematic presentations see Debeaujon et al. 2007 and Le et al.
2010). In Arabidopsis, this process occurs over a period of
1821 days, depending on growth conditions. The different
developmental periods during seed formation are assigned
as days after fertilization (DAF). As in many oilseed crops,
plastids undergo two differentiation processes during seed
formation, i.e. chloroplast formation and dedifferentiation.
Seeds become green at the beginning of maturation and it
has been proposed that photoheterotrophic plastids contribute to seed filling and seed quality by providing oxygen
and ATP that are used for respiration and biosynthesis (rev.
in Weber et al. 2005). Afterwards, chloroplasts dedifferentiate into non-photosynthetic plastids present in dry
A detailed morphological analysis of plastid differentiation and dedifferentiation in developing embryos during
seed formation has been performed by Mansfield and
Briarty (1991, 1992). They show that plastids remain
undifferentiated up to the late globular stage. In the torpedo
stage, many plastids contain already a well-developed
granal system and greening starts. However, at the end of
the maturation phase chloroplasts degenerate into reduced
non-photosynthetic plastids, called eoplasts. The maturation phase is further characterized by a reduction of plastid
number in the developing embryo that should be due to
plastid destruction. This suggests that plastids should not
be important in dry seeds and during germination, a conclusion that contradicts results showing that plastid gene
expression is important for efficient germination (Budziszewski et al. 2001; Demarsy et al. 2006). This raises the
question of how an obvious dedifferentiation/destruction of
plastids during the post-maturation period copes with a
remarkable activation of plastid genome transcription
already during stratification (Demarsy et al. 2006).
Transcription of the higher plant plastome is performed
by two principally different transcription systems that
originate from a cyanobacteria- and proteobacteria-like
endosymbiote. The cyanobacteria-like ancestor of chloroplasts has provided a eubacteria-type RNA polymerase
(PEP) whose subunits are plastid-encoded. The proteobacteria-like ancestor of mitochondria contributed with a
phage-type RNA polymerase (NEP) to the plastid transcriptional apparatus of higher plants (Filee and Forterre
2006; Liere et al. 2011). The gene has been transferred to
the nucleus during evolution and gene duplication has
probably given rise to the two nowadays existing plastid
NEPs, named RPOTp and RPOTmp, with RPOTmp being
targeted to and being active also in mitochondria (Hedtke
et al. 2000; Kuhn et al. 2009). NEPs are mainly active in


Plant Mol Biol (2013) 82:5970

the expression of housekeeping genes (Allison et al. 1996),

and they play an important role in the building-up of the
plastid transcriptional and translational apparatus during
stratification, germination and early seedling outgrowth
(Courtois et al. 2007; Demarsy et al. 2012). RPOTmp/
RPOTp double mutants are growth arrested early after
germination thus confirming the importance of basic plastid function during seed formation or early plant development (Hricova et al. 2006).
PEP plays a predominant role in the expression of
photosynthesis related genes in leaves. However, also PEP
is already present in dry seeds and active during germination (Demarsy et al. 2006). The PEP core enzyme is
composed of four different subunits, a, b, b0 and b00 , which
are encoded on the plastid genome. The activity/specificity
of the PEP core enzyme is regulated by sigma-like transcription factors (SLFs) that are nucleus-encoded. Six different sigma factors, SIG1 to SIG6, have been described
for Arabidopsis thaliana and some of them (SIG2, SIG3
and SIG5) might play a role in embryonic photosynthesis
and very early plant development. SIG2-PEP holoenzyme
transcribes specifically the tRNA-Glu gene (Kanamaru
et al. 2001) and the psaJ gene (Nagashima et al. 2004), e.g.
it might be implicated in plastid-nucleus feedback regulation via intermediates of chlorophyll biosynthesis and in
the construction of photosystem I. SIG3 is already present
in dry seeds (Privat et al. 2003; Demarsy et al. 2006)
suggesting a specific function during seed maturation or
during germination. SIG5 might have specific functions
during embryogenesis since a knock-out mutant has been
described as embryo-lethal (Yao et al. 2003) and SIG6
seems to play a more general role in early plant development (Ishizaki et al. 2005; Loschelder et al. 2006; LerbsMache 2011).
In the present article, we have analysed changes in
plastid transcriptome patterns and changes in components
of the plastid transcriptional machinery in entire seeds on
the mRNA and on the protein level all over the three
periods of seeds formation. Results provide an explanation
to understand how degenerated non-photosynthetic plastid
can efficiently regain their transcriptional activity during
germination, even already during stratification.

Plastid mRNA profiles change from embryogenesis
to maturation and from maturation to post-maturation/
Growing seeds from 1 up to 16 days after fertilization are
represented in Fig. 1. Compared to the schematic representation of changes in internal seed structure during seed

Plant Mol Biol (2013) 82:5970


Fig. 1 Developmental stages of Arabidopsis seeds (ecotype Columbia). DAF Days After Fertilization, stage 0 = dry seeds

formation (Le et al. 2010), the development is somewhat

accelerated under our growth conditions. Chlorophyll
degradation, i.e. post-maturation, starts already at DAF13.
This difference might rely on the usage of different ecotypes (Col in our studies against WS in Le et al. 2010) and/
or on different growth conditions (16/8 h light/dark cycle
in our studies against continuous light in Le et al. 2010). As
it is technically not feasible to analyse the RNA profiles for
each DAF separately, we decided to combine DAF2 to 4
seeds (embryogenesis), DAF6 to 11 (photosynthesis/
maturation) and DAF13 to 15 seeds (post-maturation/
desiccation) for our investigations. The intermediary stage
seeds, i.e. DAF5 and DAF12 seeds, were eliminated in
order to get homogeneous material. At DAF5 greening
starts and at DAF12 seeds are not yet brown.
Total RNA prepared from these three stages has been
analysed by macroarray hybridisation using a plastid specific macroarray that allows analysing all plastid mRNAs
in a relative quantitative manner (Demarsy et al. 2012).
Mean values and standard deviations for each individual
mRNA are reported in Supplemental Table 1. Figure 2
shows the changes of plastid mRNA levels over the three
developmental stages. As the diagrams are drawn to the
same scale, values are directly comparable. We note that
all photosynthesis-related mRNAs belonging to the functional groups of Photosystem I (PSI), PSII, ATPsynthase
(ATP) and electron transport (ETR) are significantly upregulated in DAF6/11 where seeds are green and photosynthetically active. However, not all mRNAs, even within
the same functional group, accumulate to equal levels. The
most abundant mRNAs at DAF6/11 are psaJ, the psbE

operon (psbE, psbF, psbL, psbJ), psbA and atpH. These

mRNAs represent already abundant mRNAs in the DAF2/4
profile. Especially to notice is the extremely high amount
of psbA mRNA in the post-maturation period. While all
other photosynthesis-related mRNAs are strongly reduced
when compared to the DAF6/11 profile, the psbA mRNA
level remains constant.
To better evaluate the mRNAs of the lowly expressed
genes, mRNAs of the same functional groups are represented on individual scales in supplemental Figure 1. This
shows that also most of the lowly produced mRNAs
increase during maturation and decrease during desiccation. However, there are some exceptions that should merit
attention. Some r-protein mRNAs (rps11, rps12, rps14,
rps19 and rpl20), the psaB, ndhK petN and accD mRNAs
as well as all three rpo mRNAs do not decrease during
desiccation, some of these mRNAs even increase. These
differences in the accumulation kinetics of individual
mRNAs during seed formation are responsible for considerable differences between plastid transcriptome patterns of the not yet (DAF2/4) and no more (DAF13/15)
photosynthetic plastids. The comparison of all three stages
of seed development shows that the strongest quantitative
changes in mRNA levels concern photosynthesis-related
genes. Quantitative changes of mRNAs coding for housekeeping functions like transcription and translation are
much less pronounced. Although very lowly expressed, the
continuous increase of all three mRNAs coding RPO proteins was intriguing and prompted us to investigate specifically on the activity and quantity of plastid RNA
polymerases in the following.



Plant Mol Biol (2013) 82:5970

Fig. 2 Changes in plastid

transcriptome profiles of DAF2/
4, DAF6/11 and DAF13/15
seeds. Mean values of plastid
mRNA from the three seed
developmental stages are
presented at the same scale.
Values have been calculated
from three independent
macroarray experiments
(biological replicates) and only
mRNAs for which standard
deviations have been lower than
50 % are presented. The
mRNAs that belong to the same
protein complex are grouped
together. RPO RNA polymerase
PEP, RP ribosomal proteins,
NDH NDH complex, ATP ATP
synthase, ETR electron transport
chain, PSI photosystem I, PSII
photosystem II, DIV other
plastid encoded proteins. Mean
values and standard deviations
for each plastid mRNA are
reported in Supplemental
Table 1

PEP and RPOTp generated transcripts follow the same

kinetics during seed formation, but RPOTmp generated
transcripts accumulate differently
In order to investigate on promoter usage and to verify
whether the activity of all three plastid RNA polymerases,
PEP, RPOTp and RPOTmp, change in the same way during
seed formation, next we analysed some selected plastid
precursor RNAs by primer extension (Fig. 3). As reminiscent for PEP transcription we have analysed the rbcL
and psbE precursor RNAs. RPOTp transcription has been
followed by revealing clpP mRNA precursor molecules. As
example for RPOTmp made transcripts we have analysed


the PC promoter initiated rRNA precursor molecules

(Courtois et al. 2007).
The two rbcL precursor RNAs correspond to the -182
primary transcript and the -59 processing product (Mullet
et al. 1985; Fig. 3, lanes 13). Both transcripts increase
strongly from DAF2/4 to DAF6/11 (lanes 1 and 2) and
decrease to a not detectable RNA level during desiccation
(lane 3), that is transcription and processing decrease
concomitantly. The psbE and clpP precursor RNAs, corresponding to positions -127 and -53 from the ATG
codon, respectively (Fig. 3, lanes 49; Hajdukiewicz et al.
1997; Demarsy et al. 2012), follow the same kinetic. No
supplementary transcript is detectable for neither of these

Plant Mol Biol (2013) 82:5970


Fig. 3 Analysis of plastid precursor RNAs by primer extension

10 lg (lanes 19) or 2 lg (lanes 1012) of total RNA have been used
for primer extension analysis using rbcL (lanes 13), psbE (lanes
46), clpP (lanes 79) and 16S rRNA (lanes 1012) specific primers.
RNA was isolated from DAF2/4 (lanes 1, 4, 7 and 10), DAF6/11
(lanes 2, 5, 8 and 11) and DAF13/15 (lanes 3, 6, 9 and 12) seeds, and
the synthesized cDNAs were analysed by separation on 6 % (w/v)

denaturing polyacrylamide gels. M molecular size standards in bp.

cDNAs are marked by arrows. The number of bases indicates the
distance of the 50 -end of the corresponding mRNAs from the ATG
translation codon. The loading control (Lo) corresponds to a 32Plabelled PCR product that is added to the primer extension mixture
before cDNA purification in order to standardize an eventual loss of
cDNA during chloroform/phenol treatment

RNAs indicating that there is no activation of alternative

promoters during seed formation. Concerning the ribosomal precursor RNAs, RPOTmp specific PC-initiated
transcripts (Courtois et al. 2007) are of very low abundance. However, they do not change remarkably during
seed formation. In contrast, PEP initiated P2 transcripts are
already present in considerable amount at DAF2/4. They
increase still during DAF6/11 and decrease tremendously
during desiccation (lanes 1012).

plastid transcriptional apparatus. For comparison, we have

also analysed some other nucleus encoded plastid related
mRNAs and some other plastid encoded mRNAs. The
analyses were done with three independently made RNA
preparations. In two of the experiments RNA levels have
been analysed by relative quantitative RT-PCR and in the
third experiment by qRT-PCR. Results are reproducible in
all three experiments and mean values could be calculated
by combining all three experiments (Fig. 4). Transcripts of
other plastid target genes and of nucleus-encoded plastidrelated mRNAs are already expressed at DAF2/4. They
increase considerably during maturation (DAF6/11) and
decrease strongly during desiccation (Fig. 4c, d). On the
other hand, mRNAs encoding PEP subunits do not change
considerably during seeds formation or increase during
desiccation. NEP encoding mRNAs (RpoTmp and RpoTp)
increase tremendously during post-maturation and in dry
seeds (Fig. 4a). Transcripts encoding sigma factors, necessary for PEP activity and specificity, increase during seed
formation with the exception of the sig1 mRNA that
decreases (Fig. 4b).
If we compare the results obtained for mRNAs of the
plastid transcriptional machinery (Fig. 4a, b) with the
expression patterns of the majority of the plastid transcripts
(Fig. 2, supplemental Table 1 and supplemental Figure 1)
and with the expression of some selected plastid transcripts
(Fig. 4c) it becomes evident that mRNAs encoding the
plastid RNA polymerases increase lightly (PEP) or augment considerably (NEP) between DAF6/11 and DAF13/
15 and in dry seeds when mRNAs made by this transcriptional machinery diminish. Other nucleus-encoded

Most of the mRNAs encoding the plastid transcriptional

machinery augment during the post-maturation period
or do not change in abundance during seed formation
After having shown that most of the plastid-encoded
mRNAs diminish strongly during post-maturation, we
wanted to answer the question of whether the expression of
components of the plastid transcriptional machinery
reflects the expression of their target genes. The array
analyses already indicated that PEP subunits encoding
transcripts accumulate differently than most of the other
plastid mRNAs during seed formation (see supplemental
Table 1 and RPO in supplemental Figure 1), but these
values are among the lowest of the array and should be
confirmed with another method.
In the following, we have used relative quantitative
RT-PCR with the 18S rRNA as internal standard (QuantumRNATM 18S Internal Standard Kit, Ambion) and
qRT-PCR with the ASAR1 coding mRNA (at4g0208) as
reference (Dekkers et al. 2012; Fig. 4) to investigate on
changes of mRNA levels coding for components of the



Plant Mol Biol (2013) 82:5970

Fig. 4 Differential changes in transcript abundance during seed

formation. Total RNAs have been purified from stage 2/4 (blue
columns), stage 6/11 (red columns), stage 13/15 (green columns) and
dry seeds (violet columns) from three biological replicates. RNAs
were analysed by relative quantitative RT-PCR using 18S RNA as
internal standard (2 biological replicates) and by qRT-PCR using
at4g02080 mRNA as internal standard (three technical replicates for
the third biological replicate). Relative quantitative RT-PCR signals
have been quantified after agarose gel separation using ImageJ
software. For each experiment, values of each mRNA have been

summed up from all four seed developmental stages, set to 100 and
RNA levels present in each developmental stage are expressed in
percentage. a Abundance of mRNAs encoding components of the
three different plastid RNA polymerases. b Analysis of mRNAs
encoding sigma-like transcription factors of PEP. c Analysis of
plastid-encoded mRNAs that are transcribed preferentially by PEP
(rbcL and psbB) or by NEP (accD). d Abundance of nucleus-encoded
mRNAs corresponding to plastid-localized photosynthesis-related

mRNAs of plastid localized proteins like RbcS, HemA and

LhcA1 (Fig. 4d) behave like most of the plastid encoded
mRNAs i.e. they diminish strongly between the maturation
and the post-maturation phases. Altogether, we conclude
from these experiments that mRNAs encoding the plastid
transcriptional machinery show principally different
expression patterns than mRNAs made by the plastid
transcriptional apparatus. In addition, the nucleus-encoded
transcripts of the plastid transcriptional machinery (RpoTp
and RpoTmp) are much differently expressed than nucleusencoded transcripts of other plastid-localized proteins.

pattern. Therefore, the amounts of other proteins are relatively diminished if the same quantity of total proteins of
the different extracts is analysed (supplemental Figure 2a).
To approach this problem, Western analyses have been
performed by loading protein quantities corresponding
always to an equal number of seeds (Fig. 5). In this way
high molecular weight proteins are present in comparable
amounts in the DAF6/11, 13/15 and dry seeds (see supplemental Figure 2b). Also, in the immuno-detection
analyses two additional protein concentrations of the 6/11
extract have been run on the same gel and have been
revealed concomitantly in the antibody reactions in order to
verify that the antibodies are in excess and that the reaction
is still in the linear range (Fig. 5a, b, left-hand side, lanes 5
and 6). The schematic presentation on the right hand side
shows the results after quantification of the signals using
ImageJ software.
In all cases, the immuno-detection analyses show a raise
of protein levels between DAF2/4 and DAF6/11. Afterwards we observe a difference for components of the
plastid transcriptional apparatus and proteins that originate
from plastid target genes. Protein levels of plastid target
genes decrease strongly during post-maturation/desiccation
and do not change any more up to dry seeds. In contrast,
the protein levels of the two phage-type plastid RNA
polymerases, RPOTp and RPOTmp, and the PEP subunit

Protein components of the plastid transcriptional

machinery augment during maturation and are
preserved during desiccation
The next question to answer was of whether proteins of
components of the plastid transcriptional apparatus and
other plastid-localized proteins accumulate in the same
manner as their corresponding mRNAs. We analysed some
of the proteins by immuno-detection. An important problem concerning the Western analyses refers to the lack of
an internal protein standard. The overall protein pattern
changes tremendously during seed formation, due to the
accumulation of high amounts of storage proteins, visible
in the low molecular weight region in the electrophoresis


Plant Mol Biol (2013) 82:5970


Fig. 5 Changes in protein levels for components of the plastid

transcriptional machinery and some other plastid proteins during seed
formation. Total proteins corresponding to an equal number of seeds
of DAF2/4 (lanes 1), DAF6/11 (lanes 2), DAF13/15 (lanes 3) and dry
seeds (0, lanes 4) were analysed after separation on 7 % denaturing
polyacrylamide gels and transfer to Nitrocellulose membranes.
Protein profiles are revealed by antibody decoration (left hand side)
and signals were quantified using ImageJ software (right hand side).
Protein loading corresponds to seeds of 10 siliques for RPOTp and
RPOTmp, to 9 siliques for RPOB, to 1.5 siliques for RBCL and to 3
siliques for all other proteins. Seeds of three siliques correspond to a
protein quantity of 1, 10, 13 and 26 lg for DAF2/4, DAF6/11,

DAF13/15 and dry seeds, respectively. Control reactions were

performed with the 6/11 protein extract by loading either 1/3 (lane
5) or threefold (lane 6) the protein concentration used in lane 2.
Except for SIG2, the Western blot experiments have been repeated
once or twice and mean values and standard deviation have been
calculated after signal quantification (right hand side). Values for
RPOB and ATPH have been calculated from three replicates and all
other proteins have been calculated from two replicates. a Proteins of
components of the plastid transcriptional apparatus. b Other proteins
made from plastid encoded genes. The star in b marks an artefactual
band that appears also with other antibodies and in the preimmune

RPOB do not change, i.e. the protein levels that are

achieved during the maturation period do not decrease but
are preserved up to the dry seed. An exception represents
the transcription initiation factor SIG2 which is known to
be important during early plant development. The SIG2
protein level decreases during post-maturation. It seems
that mainly the PEP core enzyme is preserved up to dry
seeds, but the transcription factor might need de-novo
synthesis for germination.

chloroplasts dedifferentiate into reduced non-photosynthetic plastids. Thus, seed formation represents an interesting experimental system to investigate on changes in
plastid gene expression connected to chloroplast dedifferentiation. (2) RNAs coding for plastid transcriptional
components as well as proteins of the three plastid core
RNA polymerases are already present in dry seeds
(Demarsy et al. 2006) raising the question of how they
come into the seeds. A priori, one would expect that these
components diminish during plastids dedifferentiation that
occurs in the desiccation period. To investigate on these
two problems we have analysed the expression of the
plastid genome and the expression of the plastid transcriptional machinery during seed formation on the RNA
and on the protein level.

In the present article, we have analysed the expression of
the plastid genome and the expression of components of
the plastid transcriptional machinery during seed formation. This work was initiated to investigate on two different
problems. (1) Seed development proceeds via three welldefined phases, i.e. embryogenesis, maturation and desiccation. Arabidopsis seeds develop photosynthetically
active chloroplasts and become green during the maturation phase. At the end of the maturation phase, these

Changes of plastid gene expression profiles during seed

As expected, results reveal a general increase of all plastid
mRNAs between DAF2/4 and DAF6/11 (Fig. 2, supplemental Figure 1). Photosynthesis related genes represent



already the most expressed genes during embryogenesis

(Fig. 2, DAF2/4) and their mRNA levels increase specifically with the onset of the greening period. Having analysed entire seeds, we cannot conclude on the question of
whether the expression of photosynthesis-related genes
occurs preferentially in one of the seed organs or tissues.
However, genome-wide gene expression profiling in the
different seed organs/tissues using laser-capture microdissection reveals that plastid genes are expressed in all
seed tissues, from the preglobular up to the mature green
seed stage (Belmonte et al. 2013). Our 2/4 stage comprises
the preglobular, globular and heart stage. The 6/11 stage
corresponds to the linear cotyledon up to the mature green
seed stage. In order to compare our results to those
obtained for each tissue separately, we have extracted all
plastid mRNA values from Dataset S2 of Belmonte et al.
(2013). With the exception of preglobular, globular and
heart stage embryos, plastid gene expression profiles are
very similar in all seed tissues/organs. They are also very
similar to our DAF2/4 and DAF6/11 profiles and photosynthesis related mRNAs are everywhere present at the
highest levels (supplemental Figure 3).
From the mature green seed stage up to the dry seed the
embryo occupies the largest part of the seed and our
analyses of the desiccation period and of dry seeds should
mainly reveal embryo specific mRNAs. During the phase
of plastid dedifferentiation/destruction, i.e. during desiccation, photosynthesis-related mRNAs decrease strongly,
even to lower levels than present in non-photosynthetic
DAF2/4 seeds (Fig. 2, DAF13/15). A remarkable exception
from this rule represents the tremendously high psbA
mRNA level that persists during the desiccation period
(Fig. 2). No such accumulation of psbA in non-photosynthetic plastids has been described so far, neither for chromoplasts (Valkov et al. 2009) nor for amyloplasts (Kahlau
and Bock 2008). Why psbA mRNA molecules do not
diminish in the same way as do all other PSII mRNAs
during seed desiccation? We can exclude a specific protection mechanisms of the psbA mRNA from degradation
by RNA double-strand formation (Zghidi-Abouzid et al.
2011) because the psbA anti-sense RNA level is very low
(not shown). On the other hand, we have recently shown
that stored psbA mRNA is degraded during germination
(Demarsy et al. 2012) suggesting that this mRNA constitutes a reserve of nucleotides used during germination.
From these results we suggest that the observed psbA
mRNA accumulation during seed formation might be
related to a function as nucleotide reserve for germination.
The question of whether the general diminution of most
of the plastid mRNAs between DAF6/11 and desiccation is
linked to further diminution of plastid numbers per cell
cannot be answered at present because data concerning
exact plastid numbers per cell during DAF13/15 are not


Plant Mol Biol (2013) 82:5970

available. The analyses of Mansfield and Briarty cover the

period up to 216 HAF (hours after flowering), corresponding to DAF9. In later developmental stages, for
example at DAF15, plastids are no more identifiable due to
the abundance of lipid and protein storage bodies (Ruppel
et al. 2011). It would be surely interesting in the future to
determine plastid numbers per cell also during later stages
of seed formation, for instance by using fluorescence
labelling of plastids for better visualisation (Law et al.
2012). However, the conclusion that the psbA mRNA and
mRNAs encoding components of the plastid transcriptional
machinery are specifically conserved or enriched during
desiccation and stored in dry seeds will not change if the
plastid number/cell further diminishes during desiccation.
Expression of mRNAs coding for components
of the plastid transcriptional apparatus during seeds
What may we conclude now concerning the expression of
components of the plastid transcription system during seed
formation? We observe increasing levels of RPOTmp and
RPOTp mRNAs as well as of mRNAs encoding some of
the different sigma factors during desiccation (Fig. 4a, b).
This increase could be related to specific stabilization of
these mRNAs and/or to their de novo synthesis. Only stabilization could maximally lead to equal RPOTmp and
RPOTp mRNA levels. An increase should also rely on de
novo synthesis. Gene expression even in the dry seed stage
has been described (Leubner-Metzger 2005), and it might
also occur during desiccation. However, degradation/preservation/or synthesis of nucleus-encoded plastid-related
mRNAs is mRNA-specific. For example, photosynthesis
related mRNAs are strongly down regulated during desiccation (Fig. 4d) while mRNAs coding for plastid transcriptional components are up regulated (Fig. 4a, b). The
same holds true for plastid-encoded mRNAs. While PEP
subunits encoding mRNAs and PC-initiated ribosomal
precursor RNAs (i.e. NEP transcripts) remain stable,
mRNAs encoding photosynthesis related proteins and P2initiated ribosomal precursor RNA (i.e. PEP transcripts) are
down regulated (Figs. 3, 4a, c). The difference in expression of NEP and PEP transcripts might be related to their
different turnover rates as described by Cahoon et al.
(2004) for developing maize plastids. Whatever the reason,
our results show that mRNAs encoding components of the
plastid transcriptional machinery are specifically preserved
or even newly synthesized during desiccation and that they
are stored in dry seeds.
It is known since long time that dry seeds contain a large
amount of stored, translatable nucleo/cytoplasmic mRNA
(Dure and Waters 1965). Between 12,000 and 17,000
transcripts are present in dry seeds of different plant

Plant Mol Biol (2013) 82:5970

species (Nakabayashi et al. 2005; Sreenivasulu et al. 2004;

Howell et al. 2009) and the quality of stored mRNAs is
probably important for efficient germination (Rajjou et al.
2012). Experiments using inhibitors of nuclear transcription (a-amanitin) and cytoplasmic translation (cycloheximide) have shown that, although retarded, germination still
occurs in the presence of a-amanitin, but it is abolished in
the presence of cycloheximide (Rajjou et al. 2004). This
shows the extreme importance of long-lived mRNA to
assure immediate translation and production of proteins
during germination. Regarding the plastidial transcription
system, we assume that the corresponding stored nucleusencoded mRNAs, that is RpoTp, RpoTmp and mRNAs
encoding sigma-like transcription factors, are among these
immediately translated stored mRNAs.
We suggest that the remarkable enhancement of RpoTp
and RpoTmp mRNA levels during desiccation and in dry
seeds serves providing ready to use translationally competent mRNA for cytoplasmic ribosomes in order to augment NEP protein levels immediately during germination.
The same should hold true for sigma factor encoding
mRNAs. PEP activity is regulated by sigma factor supply
and rapid de novo synthesis of sigma factor(s) from stored
mRNAs would activate stored PEP core RNA polymerase
for rRNA transcription during imbibition/stratification and
prepare PEP for transcription of photosynthesis related
genes that is considerably induced after cold release, yet
before germination is accomplished (Demarsy et al. 2012).
Expression of proteins of the plastid transcriptional
apparatus during seeds formation
Besides mRNAs, also proteins of the plastid transcriptional
apparatus are preserved during desiccation and stored in
dry seeds. RPOTp, RPOTmp and the b-subunit of PEP
(rpoB), reminiscent of the PEP core enzyme, increase
during maturation. Afterwards their protein levels do not
change any more up to dry seeds (Fig. 5a). In contrast, all
other analysed plastid proteins that are not PEP components diminish during post-maturation (Fig. 5b). An
exception represents the early sigma factor SIG2 that is
necessary for PEP activity. The SIG2 protein diminishes
during desiccation (Fig. 5a) as do all other analysed plastid
proteins that are not PEP components (Fig. 5b). Being a
regulatory protein, SIG2 has probably a higher turnover
rate than RNA polymerase core subunits. Altogether, we
interpret our results in that the two NEPs, the PEP core
enzyme and probably traces of sigma factors are preserved
during seeds formation and are stored in dry seeds. They
are still transcriptional active and their activity is necessary
for efficient germination. This can be concluded from the
facts that all three plastid RNA polymerases are already
active during stratification and germination tests in the


presence of Tagetin, a specific inhibitor of PEP transcriptional activity, show the importance of stored PEP for
efficient germination (Demarsy et al. 2006, 2012; Courtois
et al. 2007). Storage of RNA polymerases in active form in
dry seeds has recently also been shown for nuclear RNA
polymerases (Kimura and Nambara 2010).
Altogether, the kinetics of mRNA and protein accumulation during seed formation indicates that most of the here
analysed plastid proteins and plastid related mRNAs that are
present in dry seeds are probably remnants of the intermediary metabolically active photosynthetic period, that is
maturation. However, components of the plastid transcriptional apparatus are specifically preserved from degradation
during the phase of plastid degeneration/degradation (i.e.
desiccation) or even increased by de novo synthesis. The
expression patterns of these components are schematically
demonstrated in Fig. 6. These expression patterns provides
an answer to the question of how degenerated non-photosynthetic plastids in dry seeds can immediately and efficiently regain their transcriptional activity during
germination, even already during stratification.

Plant material
Arabidopsis plants have been grown on soil/vermiculite
(4/1) in 16 h light/8 h dark cycle at 80 lE/m2/s and 23 C.
Individual flowers have been labelled at opening (DAF0)
by knotting with differently coloured threads where each
colour corresponds to a given day of opening. Siliques of
the desired stages were harvested and developing seeds
were obtained from peeled siliques as described (Allorent
et al. 2010). After harvesting the developing seeds were
immediately frozen in liquid N2 and stored at -80 C.
Mature, dry seeds were stored at 16 C.
RNA isolation, relative quantitative RT-PCR
and quantitative RT-PCR (qRT-PCR)
The protocol of RNA isolation was adapted from Suzuki
et al. (2004) by introducing several modifications. Absence
of plastidial DNA contamination was regularly checked by
PCR. The procedure is described in all details by Allorent
et al. (2010).
Synthesis of cDNAs was performed in 55 ll reaction
volumes using 1 lg of total RNA, 0.5 mM dATP, dGTP,
dCTP and dTTP, 1 lg of random primers (Invitrogen), 3.7
U of RNAse inhibitor (Euromedex) and 200 U of reverse
transcriptase (Superscript II, Invitrogen) at 42 C for
50 min. The reactions were stopped by heating at 70 C for
10 min and the solutions were kept frozen at -20 C until



Plant Mol Biol (2013) 82:5970

Fig. 6 Schematic presentation

of expression profiles of
transcripts and proteins of
components of the plastid
transcriptional apparatus and of
other plastid-encoded target
genes during seed development

further usage. PCR reactions are made by using aliquots of

these cDNA solutions corresponding to the same amount of
18S rRNA.
The question of whether 18S rRNA can be used as reference mRNA also for studying seed development was
answered by comparing results obtained with the QuantumTMRNA Universal 18S kit (Ambion) with results obtained
by real-time PCR using the at4g02080 mRNA as reference.
This mRNA has recently been identified as one of the most
stable seed reference mRNA (Dekkers et al. 2012). In addition, agarose gel electrophoresis of the same quantities of total
RNA from each developmental stage shows that nucleocytoplasmic rRNA levels remain constant (not shown).
RNA quantities of different samples were adjusted to
equal amounts by using the QuantumTMRNA Universal
18S kit (Ambion). PCR reactions were performed as follows: denaturation for 5 min at 95 C, amplification (30 s
at 95, 30 s 55 C, 45 s 72 C) and termination at 72 C
for 5 min. The number of amplification cycles was determined for each of the RNA samples in pre-assays in order
to assure that the reaction is still in the exponential phase.
PCR products are finally analysed by electrophoresis in
2 % (w/v) Agarose gels in 0.59 TBE and stained with
ethidium bromide.
Real-time PCR was performed in a Rotor-Gene RG3000 thermocycler (Corbett Research, UK) using SYBRGreen for fluorescence product detection and the primers
designed for each gene (see Supplemental Table 2). Notemplate controls were joined to every PCR reaction and
melting curve analyses were carried out to prove the
specificity of the reaction using suppliers software. Genespecific mRNAs were normalized to the amount of
at4g02080 mRNA as described in Dekkers et al. (2012).
RNA analysis by primer extension
Primer extension experiments have been performed using
2 lg (16S rRNA) or 10 lg (all other mRNAs) of total


RNA. Primers have been 50 -labelled using c[32P]dATP (10

lC/ll; 3,000 Ci/mmol, Perkin Elmer, France) and T4
polynucleotide kinase (Invitrogen) according to the suppliers protocol. The reaction is stopped by denaturation of
the enzyme at 65 C for 10 min and the labelled oligonucleotide is separated from unincorporated dATP by gel
filtration through Sephadex G50 (GE Healthcare). The
labelled primer is annealed with the corresponding RNA
after previous denaturation at 65 C for 10 min by incubating at 1 C below the TM of the primer for 20 min.
Retro-transcription was performed using 100 U of Superscript II (Invitrogen) at 42 C for 50 min. The reaction was
stopped by heating at 70 C for 15 min and remaining
RNA was digested by treatment with RNaseA (Fermentas).
Before cDNAs were purified by phenol/chloroform treatment and ethanol precipitation, a 32P-labelled PCR product
was added to the reaction as loading control. The cDNAs
were separated on a 6 % polyacrylamide denaturing gel
together with a molecular size standard (1 kb ladder,
The following primers have been used for primer
extension: 50 -CCCAACACTTGCTTTAG-30 (rbcL), 50 -GA
CDNA synthesis and array hybridization
CDNA synthesis
cDNAs have been synthesized as previously described
(Zghidi et al. 2007; Demarsy et al. 2012). Total RNA
preparations were carefully verified for the absence of
plastid DNA. 4 lg of each RNA preparation have been
labelled for macroarray hybridisation. RNA was reverse
transcribed using a primer mix that contains specific primers
corresponding to the 80 protein coding genes and their antisense RNAs that can be analysed on the macroarray

Plant Mol Biol (2013) 82:5970


(Uniplastomic). Sense-primers are localized as near as

possible to the 30 -end of the 60-mer oligonucleotides that
have been spotted on the macroarray. Anti-sense primers
have been chosen in the 50 -UTR or at the 50 -end of the
mRNA. All primers have very close TMs. The primer concentration is in excess over the concentration of the specific
mRNAs as tested by comparing array hybridisation patterns
made with cDNAs of two different RNA concentrations (e.g.
2 and 4 lg of RNA used for reverse transcription). Reactions
were performed in the presence of 100 lCi of a32P dATP
(Perkin Elmer, France) using Superscript II Reverse Transcriptase (Invitrogen). Samples were treated with RNase H
(Fermentas, France) at 37 C for 15 min and non-incorporated deoxyribonucleotides were removed by passage
through Sephadex G50. An aliquot of each of the synthesized cDNAs is always analysed on a 6 % denaturing
polyacrylamide gel in order to verify the quality of the
synthesized cDNA.

ECL? detection kit (GE Healthcare) using Typhoon 9400

(GE Healthcare). Antibodies were obtained from Agrisera
(anti-RBCL, anti-PSBB, anti-ATPB and anti-ATPH) and
from Uniplastomic (anti-ACCD, anti-RPS7 and antiRPOB). SIG2, RPOTmp and RPOTP antibodies have been
purified on the corresponding peptides before usage as previously described (Azevedo et al. 2006). The secondary
antibody was obtained from Interchim (Peroxydase goat
anti-rabbit IgG, H ? L).

Array hybridization

Allison LA, Simon LD, Maliga P (1996) Deletion of rpoB reveals a

second distinct transcription system in plastids of higher plants.
EMBO J 15:28022809
Allorent G, Lambert E, Lerbs-Mache S, Courtois F (2010) RNA
isolation from developing Arabidopsis thaliana seeds suitable for
gene expression analyses. J Endocytobiosis Cell Res 20:2633
Azevedo J, Courtois F, Lerbs-Mache S (2006) Sub-plastidial localization of two different phage-type RNA polymerases in spinach
chloroplasts. Nucleic Acids Res 34:436444
Belmonte MF, Kirikbride RC, Stone SL et al (2013) Comprehensive
developmental profiles of gene activity in regions and subregions
of the Arabidopsis seed. Proc Natl Acad Sci USA 14:E435E444
Budziszewski GJ, Lewis SP, Glover LW, Reineke J, Jones G, Schlate
Ziemnik L, Lonowski J, Nyfeler B, Aux G, Zhou Q, McElver J,
Patton DA, Martienssen R, Grossniklaus U, Ma H, Law M,
Levin JZ (2001) Arabidopsis genes essential for seedling
viability: isolation of insertional mutants and molecular cloning.
Genetics 159:17651778
Cahoon AB, Harris FM, Stern DB (2004) Analysis of developing
maize plastids reveals two mRNA stability classes correlating
with RNA polymerase type. EMBO Rep 5:801806
Courtois F, Merendino L, Demarsy E, Mache R, Lerbs-Mache S
(2007) Phage-type RNA polymerase RpoTmp transcribes the rrn
operon from the PC promoter at early developmental stages in
Arabidopsis. Plant Physiol 145:712721
Debeaujon I, Lepiniec L, Pourcel L and Routaboul JM (2007) Seeds
coat development and dormancy. In: Bradford K, Nonogaki H
(eds) Seed development, dormancy and germination. Blackwell
Publishing vol. 27, pp 2443
Dekkers BJW, Willems L, Bassel GW, van Bolderen-Veldkamp RP,
Ligterink W, Hilhorst HWM, Bentsink L (2012) Identification of
reference genes for RT-qPCR expression analysis in Arabidopsis
and tomato seeds. Plant Cell Physiol 53:2837
Demarsy E, Courtois F, Azevedo J, Buhot L, Lerbs-Mache S (2006)
Building-up of the plastid transcriptional machinery during
germination and early plant development. Plant Physiol
Demarsy E, Buhr F, Lambert E, Lerbs-Mache S (2012) Characterization of the plastid-specific germination and seedling establishment transcriptional programme. J Exp Bot 63:925939
Dure L, Waters L (1965) Long-lived messenger RNA: evidence from
cotton seed germination. Science 147:410412

Prehybridisation (1 h at 65 C) and hybridisation (3 days at

65 C) were performed in 0.5 M NaHPO4, pH 7.2; 1 mM
EDTA; 7 % SDS and 1 % BSA as previously described
(Zghidi et al. 2007; Demarsy et al. 2012). After hybridisation, macroarrays were washed in 40 mM NaHPO4, pH
7. 2.1 mM EDTA and 7 % SDS at room temperature for
10 min followed by 65 C washing for 5 min. After
12 weeks of exposure to Fujifilm Imaging Plates, the
plates were analysed using a Phosphoimager (Fujifilm
FLA-8000) and the accompanying software (ArrayGauge).
Protein analyses
Total protein was extracted from developing seeds in eppendorf tubes by 3 cycles of freezing/thawing in liquid
nitrogen and grounding during thawing after addition of
100 ll of extraction buffer (62.5 mM Tris/HCl pH 6.8,
2.5 % (w/v) SDS, 2 % (w/v) DTT, 10 % (v/v) Glycerol).
After centrifugation at 160009g for 15 min the supernatant
was recovered and the pellet was again grounded with
50 ll extraction buffer. After centrifugation of the second
extraction the supernatants were combined and cleaned by
re-centrifugation at 160009g for 25 min. Proteins were
precipitated from the final supernatant by adding 5 volumes
of acetone.
Protein concentrations have been determined using the
protocol of Esen (1978) and equal quantities of protein have
been used for immunoblot analyses of specific proteins. Gel
blot analyses of proteins were performed after separation on
12.5 % denaturing polyacrylamide gels (Laemmli 1970).
After transfer onto Nitrocellulose (0.45 lm, Whatman),
proteins were revealed after immunodecoration using the

Acknowledgments Financial support was obtained from the French

agency of research (GENOPLANTE project ANR-07-GPLA-013001), the Centre National de la Recherche Scientifique and the
Ministe`re de lEnseignement Superieur et de la Recherche. The
authors declare that they have no conflict of interest and they thank D.
Job for critical reading of the manuscript.



Esen A (1978) A simple method for quantitative, semiquantitative and
qualitative assay of protein. Anal Biochem 89:264273
Filee J, Forterre P (2006) Viral proteins functioning in organelles: a
cryptic origin? Trends Microbiol 13:510513
Hajdukiewicz PTJ, Allison LA, Maliga P (1997) The two RNA
polymerases encoded by the nuclear and the plastid compartment
transcribe distinct groups of genes in tobacco plastids. EMBO J
Hedtke B, Borner T, Weihe A (2000) One RNA polymerase serving
two genomes. EMBO rep 11:435440
Howell KA, Narsai R, Carroll A, Ivanova A, Lohse M, Usadel B,
Millar AH, Whelan J (2009) Mapping metabolic and transcript
temporal switches during germination in rice highlights specific
transcription factors and the role of RNA instability in the
germination process. Plant Physiol 149:961980
Hricova A, Quesada V, Micols JL (2006) The SCABRA3 nuclear gene
encodes the plastid RpoTp RNA polymerase, which is required
for chloroplast biogenesis and mesophyll cell proliferation in
Arabidopsis. Plant Physiol 141:942956
Ishizaki Y, Tsunoyamz Y, Hatano K, Ando K, Kato K, Shinmyo A,
Kobori M, Takeba G, Nakahira Y, Shiina T (2005) A nuclearencoded sigma factor, Arabidopsis SIG6, recognizes sigma-70
type chloroplast promoters and regulated early chloroplast
development in cotyledons. Plant J 42:133144
Kahlau S, Bock R (2008) Plastid transcriptomics and translatomics of
tomato fruit development and chloroplast-to-chromoplast differentiation: chromoplast gene expression largely serves the
production of a single protein. Plant Cell 20:856874
Kanamaru K, Nagashima A, Fujiwara M, Shimada H, Shirano Y,
Nakabayashi K, Shibata D, Tanaka K, Takahashi H (2001) An
Arabidopsis sigma factor (SIG2)-dependent expression of plastid-encoded tRNAs in chloroplasts. Plant Cell Physiol 42:1034
Kimura M, Nambara E (2010) Stored and neosynthesized mRNA in
Arabidopsis seeds: effects of cyclohiximide and controlled
deterioration treatment on the resumption of transcription during
imbibition. Plant Mol Biol 73:119129
Kuhn K, Richter U, Meyer EH, Delannoy E, de Longevialle AF,
OToole N, Borner T, Millar AH, Small ID, Whelan J (2009)
Phage-type RNA polymerase RPOTmp performs gene specific
transcription in mitochondria of Arabidopsis thaliana. Plant Cell
Laemmli UK (1970) Cleavage of structural proteins during the
assembly of the head of bacteriophage T4. Nature 227:680685
Law SR, Narsai R, Taylor NL, Delannoy E, Carrie C, Giraud E,
Millar AH, Small I, Whelan J (2012) Nucleotide and RNA
metabolism prime translational initiation in the earliest events of
mitochondrial biogenesis during Arabidopsis germination. Plant
Physiol 158:16101627
Le BH, Cheng C, Bui AQ, Wagmaister JA, Henry KF, Pelletier J,
Kwong L, Belmonte M, Kirkbride R, Horvath S, Drews GN,
Fischer RL, Okamuro JK, Harada JJ, Goldberg RB (2010)
Global analysis of gene activity during Arabidopsis seed
development and identification of seed-specific transcription
factors. Proc Natl Acad Sci USA 107:80638070
Lerbs-Mache S (2011) Function of plastid sigma factors in higher
plants: regulation of gene expression or just preservation of
constitutive transcription? Plant Mol Biol 76:235249
Leubner-Metzger G (2005) b-1,3-glucanase gene expression in lowhydrated seeds as a mechanism for dormancy release during
tobacco after-ripening. Plant J 41:133145


Plant Mol Biol (2013) 82:5970

Liere K, Weihe A, Borner T (2011) The transcription machineries of
plant mitochondria and chloroplasts: composition, function, and
regulation. J Plant Physiol 168:13451360
Loschelder H, Schweer J, Link B, Link G (2006) Dual temporal role
of plastid sigma factor 6 in Arabidopsis development. Plant
Physiol 142:642650
Mansfield SG, Briarty LG (1991) Early embryogenesis in Aranidopsis
thaliana. II. The developing embryo. Can J Bot 69:461476
Mansfield SG, Briarty LG (1992) Cotyledon cell development in
Arabidopsis thaliana during reserve deposition. Can J Bot 70:
Mullet JE, Orozco EM, Chua NH (1985) Multipe transcripts for
higher plant rbcL and atpB genes and localization of the
transcription initiation sites of the rbcL gene. Plant Mol Biol
Nagashima A, Hanaoka M, Motohashi R, Seki M, Shinozaki K,
Kanamaru K, Takahashi H, Tanaka K (2004) DNA microarray
analysis of plastid gene expression in an Arabidopsis mutant
deficient in a plastid transcription factor sigma, SIG2. Biosci
Biotechnol Biochem 68:694704
Nakabayashi K, Okamoto M, Koshiba T, Kamiya Y, Nambara E
(2005) Genome-wide profiling of stored mRNA in Arabidopsis
thaliana seed germination: epigenetic and genetic regulation of
transcription in seed. Plant J 41:697709
Privat I, Hakimi MA, Buhot L, Favory JJ, Lerbs-Mache S (2003)
Characterization of Arabidopsis plastid sigma-like transcription
factors SIG1, SIG2 and SIG3. Plant Mol Biol 55:385399
Rajjou L, Gallardo K, Debeaujon I, Vandekerckhove J, Job C, Job D
(2004) The effect of a-amanitin on the Arabidopsis seed
proteome highlights the distinct roles of stored and neosynthesizes mRNAs during germination. Plant Physiol 134:15981613
Rajjou L, Duval M, Gallardo K, Catusse J, Bally J, Job C, Job D
(2012) Seeds germination and vigor. Annu Rev Plant Biol
Ruppel NJ, Logsdon CA, Whippo CW, Inoue K, Hangarter RP (2011)
A mutation in Arabidopsis SEEDLING PLASTID DEVELOPMENT affects plastid differentiation in embryo-derived tissues
during seedling growth. Plant Physiol 155:342353
Sreenivasulu N, Altschmied L, Radchuk V, Gubatz S, Wobus U,
Weschke W (2004) Transcript profiles and deduced changes of
metabolic pathways in maternal and filial tissues of developing
barley grains. Plant J 37:539553
Suzuki Y, Kawazu T, Koyama H (2004) RNA isolation from siliques,
dry seeds, and other tissues of Arabidopsis thaliana. Biotechniques 37:542544
Valkov VT, Scotti N, Kahlau S, MacLean D, Grillo S, Gray J, Bock
R, Cardi T (2009) Genome-wide analysis of plastid gene
expression in potato leaf chloroplasts and tuber amyloplasts:
transcriptional and posttranscriptional control. Plant Physiol
Weber H, Borisjuk L, Wobus U (2005) Molecular physiology of
legume seed development. Ann Rev Plant Biol 56:253279
Yao J, Roy-Chowdhury S, Allison LA (2003) AtSig5 Is an essential
nucleus-encoded Arabidopsis r-like factor. Plant Physiol
Zghidi W, Merendino L, Cottet A, Mache R, Lerbs-Mache S (2007)
Nucleus-encoded plastid sigma factor SIG3 transcribes specifically the psbN gene in plastids. Nucleic Acids Res 35:455464
Zghidi-Abouzid O, Merendino L, Buhr F, Ghulam MM, Lerbs-Mache S
(2011) Characterization of plastid psbT sense and antisense RNAs.
Nucleic Acids Res 39:53795387