A Mutant Impaired in the Production of Plastome-Encoded
Proteins Uncovers a Mechanism for the Homeostasis of
Isoprenoid Biosynthetic Enzymes in Arabidopsis Plastids
U´rsula Flores-Pe ´rez,a,b,1Susanna Sauret-Gu ¨eto,b,1,2Elisabet Gas,a,bPaul Jarvis,c
and Manuel Rodrı ´guez-Concepcio ´na,b,3
aDepartament de Gene `tica Molecular de Plantes, Centre for Research on Agricultural Genomics,
08034 Barcelona, Spain
bDepartament de Bioquı ´mica i Biologia Molecular, Universitat de Barcelona, 08028 Barcelona, Spain
cDepartment of Biology, University of Leicester, Leicester LE1 7RH, United Kingdom
The plastid-localized methylerythritol phosphate (MEP) pathway synthesizes the isoprenoid precursors for the production
of essential photosynthesis-related compounds and hormones. We have identified an Arabidopsis thaliana mutant, rif1, in
which posttranscriptional upregulation of MEP pathway enzyme levels is caused by the loss of function of At3g47450, a gene
originally reported to encode a mitochondrial protein related to nitric oxide synthesis. However, we show that nitric oxide is
not involved in the regulation of the MEP pathway and that the encoded protein is a plastid-targeted homolog of the Bacillus
subtilis YqeH protein, a GTPase required for proper ribosome assembly. Consistently, in rif1 seedlings, decreased levels of
plastome-encoded proteins were observed, with the exception of ClpP1, a catalytic subunit of the plastidial Clp protease
complex. The unexpected accumulation of ClpP1 in plastids with reduced protein synthesis suggested a compensatory
mechanism in response to decreased Clp activity levels. In agreement, a negative correlation was found between Clp
pathway enzymes might be a mechanism used by individual plastids to finely adjust plastidial isoprenoid biosynthesis to their
functional and physiological states.
Plastids are arguably the most distinctive organelles of plant
cells. Besides the central importance of chloroplasts for photo-
synthesis, and therefore for life on earth as we know it, plastids
harbor essential metabolic pathways that occur uniquely in plants
among eukaryotes. A good example is the 2-methylerythritol
4-phosphate (MEP) pathway for the biosynthesis of isoprenoid
precursors (Rodrı ´guez-Concepcio ´n and Boronat, 2002; Eisenreich
et al., 2004). Isoprenoids with primary (essential) roles in cell
architecture, respiration, and regulation of growth and develop-
ment are synthesized in all living organisms, but plant cells also
produce an astonishing diversity of isoprenoid compounds for
photosynthesis-related processes and as secondary metabo-
lites that influence interactions with their environment. Most
organisms have only one of the two currently known pathways
for the biosynthesis of the prenyl diphosphate precursors of all
isoprenoids, isopentenyl diphosphate (IPP) and its isomer dime-
thylallyl diphosphate (DMAPP). Thus, IPP and DMAPP are made
exclusively from mevalonic acid (MVA) in archaebacteria, fungi,
and animals, whereas most eubacteria (including cyanobacteria,
the ancestors of plant plastids) only use the MEP pathway. By
contrast, plants use both the MVA and MEP pathways, although
in different cellular compartments (Lichtenthaler, 1999). Sterols,
brassinosteroids, triterpenes, some sesquiterpenes, polyter-
penes, and dolichol are formed from cytosolic prenyl diphos-
phates derived from the MVA pathway. On the other hand, the
plastid-localized MEP pathway synthesizes the precursors for
photosynthesis-related compounds (carotenoids and the side
chain of chlorophylls, tocopherols, phylloquinones, and plasto-
quinones), hormones (gibberellins and abscisic acid), isoprene,
monoterpenes, and some sesquiterpenes (Figure 1). Although
some exchange of prenyl diphosphates can take place between
the cytosol and the plastid in at least some plants, including
Arabidopsis thaliana (Kasahara et al., 2002; Nagata et al., 2002;
Laule et al., 2003), the block of one of the two pathways in
seedlings cannot be rescued with common isoprenoid precur-
sors synthesized by the other pathway under normal growth
conditions (Gutierrez-Nava et al., 2004; Rodrı ´guez-Concepcio ´n
et al., 2004). Pathway- and compartment-specific regulatory
mechanisms, therefore, must be present in plant cells to ensure
that isoprenoid precursors are supplied when needed in each
Plant MEP pathway enzymes are encoded by nuclear genes
and imported into plastids (Rodrı ´guez-Concepcio ´n and Boronat,
1These authors contributed equally to this work.
2Current address: Department of Cell and Developmental Biology, John
Innes Centre, Norwich NR4 7UH, UK.
3Address correspondence to email@example.com.
The author responsible for distribution of materials integral to the
findings presented in this article in accordance with the policy described
in the Instructions for Authors (www.plantcell.org) is: Manuel Rodrı ´guez-
Concepcio ´n (firstname.lastname@example.org).
WOnline version contains Web-only data.
The Plant Cell, Vol. 20: 1303–1315, May 2008, www.plantcell.org ª 2008 American Society of Plant Biologists
2002; Eisenreich et al., 2004). 1-Deoxyxylulose 5-phosphate
(DXP) synthase (DXS) catalyzes the first reaction of the pathway,
the production of DXP from the central metabolic intermediates
glyceraldehyde 3-phosphate and pyruvate. MEP is synthesized
from DXP in the next step of the pathway, catalyzed by DXP
reductoisomerase (DXR). Several enzymatic steps transform
MEP into 1-hydroxy 2-methylbutenyl 4-diphosphate (HMBPP),
which is finally converted by the enzyme HMBPP reductase
(HDR) into a mixture of IPP and DMAPP (Figure 1). DXS, DXR,
and HDR activities have been shown to share control over the
metabolic flux of the MEP pathway (Este ´vez et al., 2001;
Mahmoud and Croteau, 2001; Botella-Pavı ´a et al., 2004; Enfissi
et al., 2005; Carretero-Paulet et al., 2006). Besides the coarse
control exerted by changes in the expression of genes encoding
mental, and metabolic signals (Rodrı ´guez-Concepcio ´n, 2006),
recent reports have demonstrated that enzyme levels are regu-
lated at posttranscriptional levels (Guevara-Garcia et al., 2005;
Sauret-Gu ¨eto et al., 2006). The molecular mechanisms involved
in such regulation, however, remain to be established.
Growth of Arabidopsis seedlings in the presence of fosmido-
mycin (FSM), a strong competitive inhibitor of DXR (Figure 1),
results in a specific block in the biosynthesis of MEP-derived
plastid isoprenoids such as chlorophylls and carotenoids (re-
ing a bleached phenotype and a developmental arrest that can
be rescued by upregulating DXS and/or DXR levels (Rodrı ´guez-
Concepcio ´n et al., 2004; Guevara-Garcia et al., 2005; Carretero-
Paulet et al., 2006; Sauret-Gu ¨eto et al., 2006). Our screening for
Arabidopsis rif (for resistant to inhibition by FSM) lines able to
develop in the presence of this inhibitor resulted in the unex-
pected isolation of several pale mutants, including rif10 (Sauret-
Gu ¨eto et al., 2006). Impaired plastid RNA processing in rif10
plants and reduced protein synthesis in chloroplasts resulted in
a posttranscriptional accumulation of DXS and DXR proteins
and FSM resistance (Sauret-Gu ¨eto et al., 2006). By contrast,
other processes affecting plastid development and causing a
pale phenotype did not result in FSM resistance (Sauret-Gu ¨eto
et al., 2006). The link between an altered production of plastid
proteins and a posttranscriptional accumulation of nucleus-
encoded MEP pathway enzymes in plastids has not been ex-
A number of albino, pale, and variegated mutants with defects
in chloroplast development have been identified that affect a
variety of plastid functions, mostly related to import and pro-
cessing of nucleus-encoded proteins, expression of the plastid
genome (plastome), and photosynthesis (Lo ´pez-Juez, 2007).
Additionally, results from mutant screens suggest that a large
number of genes with unknown function or unsuspected plastid
relevance still remain to be identified as essential for chloroplast
development (Budziszewski et al., 2001). An arrest in plastid
by FSM treatment or in mutants with defective biosynthetic
genes, resulting in proplastid-like organelles with rudimentary
thylakoids and an accumulation of vesicle structures, very low
levels of photosynthetic pigments, and little or no expression of
nuclear and plastidial genes required for chloroplast function
(Mandel et al., 1996; Nagata et al., 2002; Gutierrez-Nava et al.,
2004). Our previous results (Sauret-Gu ¨eto et al., 2006) provided
strong evidence that proteins encoded by plastidial genes
might in turn modulate the accumulation of flux-controlling
MEP pathway enzymes within plastids. Plastome-encoded pro-
teins include components of the plastidial gene expression
machinery and photosynthetic apparatus and a few other poly-
peptides, including one of the catalytic subunits of the stromal
Clp protease complex (Wakasugi et al., 2001; Adam et al., 2006).
To gain a deeper insight into the mechanisms that regulate
the levels of flux-controlling MEP pathway enzymes, we have
characterized another FSM-resistant mutant with a pale pheno-
type, rif1. Mutant rif1 plants show a posttranscriptional upregu-
lation of DXS and DXR caused by the loss of function of the
At3g47450 gene, originally reported to encode a mitochondrial
protein related to nitric oxide synthesis. However, we demon-
strate here that RIF1 is a plastidial protein and that nitric oxide is
not involved in the regulation of the MEP pathway. Our data
indicate that the RIF1 protein is most likely required for plastid
ribosome assembly, confirming that defective expression of the
plastid genome eventually results in the upregulation of MEP
pathway enzyme levels. We also show that the mechanism
responsible for such upregulation involves the stromal Clp pro-
tease complex and protein degradation within the plastid.
Loss of Function of the At3g47450 Gene Causes a
Posttranscriptional Upregulation of MEP
As reported for rif10 (Sauret-Gu ¨eto et al., 2006), mutant rif1
seedlings show pale cotyledons and a clearly delayed develop-
ment and greening of true leaves compared with the Columbia
(Col) wild type (Figures 2A to 2F), eventually resulting in smaller
plants with a characteristic virescent phenotype of pale young
leaves (those in the inner whorls of the rosette) but green mature
Figure 1. Isoprenoid Biosynthesis in Plastids.
Dashed arrows represent multiple enzymatic steps. ABA, abscisic acid;
GAP, glyceraldehyde 3-phosphate. MEP pathway enzymes (in boldface)
are DXS, DXR, and HDR. The step inhibited by FSM is indicated.
1304The Plant Cell
resistance to FSM (see Supplemental Figure 1 online). As a
result, most rif1 seedlings remained virtually unaltered in the
presence of FSMat concentrations that are lethal for the Colwild
type (Figures 2A to 2D). Real-time quantitative RT-PCR exper-
iments and immunoblot analyses showed that the FSM-resistant
phenotype of rif1 seedlings likely results from the accumulation
of increased levels of DXS (1.5-fold) and DXR (almost 2-fold)
proteins in mutant plastids without changes in gene expression
Backcrossing of homozygous rif1 plants with the Col wild type
followed by analysis of the offspring showed that all of the
phenotypes described above were recessive and linked to the
Figure 2. Phenotypes of rif1 Plants and Complemented Lines.
Wild-type (Col) and transgenic lines were germinated on MS plates supplemented or not with 50 mM FSM and grown under long-day conditions. Plants
grown in the absence of inhibitor for 15 d were then transferred to soil and grown under long days until completing their life cycle. Representative
individuals at different developmental stages are shown. Panels in each row are to the same scale.
(A) to (D) Five-day-old Col ([A] and [B]) and rif1 ([C] and [D]) seedlings grown without FSM ([A] and [C]) or with FSM ([B] and [D]).
(E) and (F) Ten-day-old Col (E) and rif1 (F) plants grown on MS plates.
(G) to (I) One-month-old plants from Col (G), rif1 (H), and a transgenic rif1 line constitutively overexpressing a RIF1-GFP fusion protein (I) grown on soil.
(J) to (L) Young leaves from the complemented mutant line shown in (I) were used for confocal microscopy to detect the green fluorescence of RIF1-
GFP (J) and the red autofluorescence of chlorophyll (K) in the same area of the leaf. Images were merged to show overlapping green and red
fluorescence in yellow (L). Bars ¼ 50 mm.
MEP Pathway Regulation by Clp Protease1305
presence of the T-DNA used to generate the activation-tagging
lines (Weigel et al., 2000). A wild-type phenotype was observed
in all F1 individuals resulting from the cross of homozygous rif1
and rif10 plants, indicating that they were not alleles but cor-
sequences in the rif1 genome showed that the T-DNA was
insertion alleles identified in the Salk collection (SALK_047882;
herein named rif1-2) and the Cold Spring Harbor Laboratory Ds-
GeneTrap lines (GT6235; herein named rif1-3) displayed the
same phenotypes reported for the original rif1 mutant (renamed
rif1-1), including a developmental delay, pale green cotyledons,
and FSM resistance (Figure 4B). Furthermore, transformation
of homozygous rif1-1 plants with a construct to constitutively
express the full-length At3g47450 cDNA fused to green fluores-
cent protein (P35S:RIF1-GFP) completely restored the wild-type
phenotype (Figures 2I and 4B), confirming that all of the distinc-
tivephenotypes described hereforrif1areindeed dueto theloss
of function of this gene.
The protein encoded by the At3g47450 gene was originally
reported to function as a nitric oxide (NO) synthase (NOS1) in
Arabidopsis (Guo et al., 2003), but concerns about the proposed
synthase activity of the protein led to its later being renamed
NOA1 for NITRIC OXIDE–ASSOCIATED1 (Crawford et al., 2006;
Zemojtel et al., 2006a). The pale phenotype and delayed growth
by treating seedlings with sodium nitroprusside (SNP), a NO
donor (Guo et al., 2003). If defective production of NO was also
responsible for the enhanced accumulation of active DXS and
DXR proteins detected in the mutant, it was expected that
treatment of rif1 seedlings with SNP would result in wild-type
levels of these enzymes and FSM sensitivity. Growth of rif1-1
seedlings on SNP-supplemented plates resulted in a partial
changes were detected in the resistance to FSM (Figure 5A) or
the levels of DXR protein (Figure 5B) compared with untreated
controls or seedlings treated with the same concentration of
sodium ferrocyanide (SFC; a SNP analog that does not produce
NO). FSM resistance of the rif1-1 line was higher than that of Col
Higher concentrations of SNP or SFC negatively influenced
seedling growth. These results indicate that the posttranscrip-
by a defect in the production of NO.
The RIF1 Protein Is Likely Required for Ribosome
Function in Plastids
RIF1 bears similarity to P-loop GTP binding proteins of the YlqF/
YawG family containing a circularly permuted GTPase domain
(Leipe et al., 2002). Comparison of representative members of
the different subfamilies described for the YlqF/YawG family and
homologous Arabidopsis proteins (Figure 4C; see Supplemental
Figure 2 online) showed that RIF1 is most similar to the Bacillus
subtilis YqeH protein, recently shown to be required for the
correct formation of the bacterial 70S ribosome and the assem-
bly or stability of the small (30S) ribosomal subunit (Uicker et al.,
2007). Overall similarity and identity percentages are relatively
Figure 3. Analysis of Gene Expression and Protein Levels of MEP
Pathway Enzymes in rif1 Seedlings.
RNA and protein were extracted from 5-d-old Col and rif1 seedlings
grown on MS plates under long-day conditions.
(A) Real-time quantitative RT-PCR analysis of transcript levels of the
indicated genes in wild-type and mutant seedlings. Levels are repre-
sented relative to those in Col seedlings and correspond to the mean and
SD from duplicate PCR analyses of two independent experiments.
(B) Immunoblot analysis with antibodies against DXS and DXR. The
position of the DXR protein (Sauret-Gu ¨eto et al., 2006) is indicated with
an arrowhead. The other major band recognized by the anti-DXR serum
is shown as a protein-loading control. Coomassie blue (C-Blue) staining
was also used to monitor total protein loading. The arrow marks the
position of the RBCL protein.
(C) Quantification of DXS and DXR protein levels in Col and rif1 seedlings
from immunoblot band intensity. Levels were normalized to those of the
unspecific band recognized by the anti-DXR serum and are represented
relative to the level in Col. Mean and SE from duplicate blots of three
independent experiments are represented.
1306The Plant Cell
low (23 and 33%, respectively), but they are higher when only
conserved domains are considered (46% similar at the GTP
binding domain and 39% similar at the putative Zn binding
domain; see Supplemental Figure 2 online). The RIF1 protein
contains an N-terminal domain that is absent from the bacterial
YqeH protein and shows features of organellar targeting pep-
tides (see Supplemental Figure 2 online). A GFP-tagged version
of the full-length Arabidopsis RIF1/NOS1/NOA1 protein was
previously found to be targeted to mitochondria in roots of stably
transformed plants (Guo and Crawford, 2005). Using a similar
P35S:RIF1-GFP construct, a wild-type phenotype (including
FSM sensitivity) was fully restored in transgenic rif1-1 seedlings
(Figure 4B) and adult plants (Figures 2G to 2I). However, we were
unable to detect any green fluorescence from the biologically
active RIF1-GFP fusion protein in mitochondria of roots of any of
the transgenic lines generated. Analysis of photosynthetic tis-
sues (cotyledons and leaves) also failed to reveal GFP fluores-
cence in mitochondria, but a clear signal was found in organelles
identified as chloroplasts because of their size and chlorophyll
was efficiently imported into isolated wild-type chloroplasts in
vitro (Figure 4D), demonstrating that plastid targeting is an
intrinsic feature of this protein and not an artifact caused by
overexpression and/or GFP fusion.
To investigate whether RIF1 and YqeH were functional homo-
was required to rescue the rif1 phenotype, a sequence encoding
the bacterial protein fused to GFP was cloned in-frame with the
plastid-targeting peptide of the MEP pathway HDS/GCPE pro-
tein (Querol et al., 2002) and constitutively expressed in mutant
rif1-1 plants. As shown in Figure 4B and Supplemental Figure 3
online, the plastid-localized P-YqeH-GFP protein was able to
Figure 4. Analysis of the Gene Mutated in rif1 and the Encoded Protein.
(A) Map of the coding region of the At3g47450 gene showing the transcription start (arrow), the exons (boxes), and the positions of the T-DNA in rif1-1,
rif1-2, and rif1-3 mutants. Untranslated region sequences are represented in gray. Bar ¼ 0.1 kb.
(B) Chlorophyll autofluorescence of 6-d-old seedlings of the indicated genotype grown on MS plates with (þ) or without (?) 50 mM FSM. Representative
individuals of transgenic rif1-1 plants constitutively expressing the fusion protein RIF1-GFP (RIF1G) or a plastid-targeted bacterial YqeH protein fused to
GFP (PYG) are included. All panels are to the same scale.
(C) Phylogenetic tree of putative GTPases of the YlqF/YawG family identified in Arabidopsis. Representative members of each proposed subfamily
(Leipe et al., 2002) are also included: B. subtilis YqeH and YlqF, Escherichia coli YjeQ, Methanococcus jannaschii MJ1464, and Schizosaccharomyces
(D) In vitro import of DXR and RIF1 into wild-type chloroplasts. Import of in vitro-translated labeled DXR (used as a positive control) and RIF1 preproteins
(positions marked with asterisks) was allowed to proceed for 10 min, and then the samples were either treated (þ) or not (?) with thermolysin to remove
nonimported proteins and analyzed by SDS-PAGE and fluorography. An aliquot of the input translation mixture was also included. The positions of the
mature proteins (i.e., processed to remove the transit peptide after import into chloroplasts) are marked with arrowheads.
MEP Pathway Regulation by Clp Protease 1307
complement the greening and growth defects of rif1-1 seedlings
and to restore FSM sensitivity. These results confirm that the
Defective Development of Plastids in rif1 Seedlings
Correlates with a General Decrease in Plastome-Encoded
The results shown above indicate that the main biological func-
tion of RIF1 takes place in plastids, presumably in ribosome
assembly. As a first approach to investigate whether plastid
function was altered in the mutant, ultrathin sections of cotyle-
under long-day conditions for 3 d were examined using trans-
mission electron microscopy (Figure 6). Etioplasts with a char-
acteristic prolamellar body were observed in etiolated Col
seedlings (Figure 6A) but not in rif1 seedlings, which only har-
boredsmallroundedplastids withplastoglobuli-like vesiclesthat
membranous structures (Figures 6B and 6C). When compared
withthe wildtype,chloroplasts of light-grown rif1seedlingswere
smaller, more irregularly shaped, and showed a general reduc-
tion in the development of thylakoid membranes and granal
chloroplasts (see Supplemental Figure 4 online). By contrast,
mitochondrial ultrastructure showed no apparent differences
between Col and rif1 seedlings in any of the tissues or growing
These results demonstrate that defects in the development and
differentiation of different types of plastids (etioplasts and chlo-
roplasts), but not mitochondria, result from the loss of RIF1
Assuming that the main role of RIF1 in Arabidopsis plastids is
related to its ability to influence ribosome activity, as suggested
by complementation analysis (Figure 4B; see Supplemental
Figure 3 online), it is predicted that altered ribosome function in
rif1 mutant plastids could result in defects in protein translation,
eventually leading to decreased levels of plastome-encoded
proteins. Consistently, the levels of the ribulose-1,5-bisphos-
phate carboxylase/oxygenase large subunit (RBCL) detected by
Coomassie blue staining after SDS-PAGE of protein extracts
were much lower in rif1 than in Col seedlings (Figures 3 and 7).
Immunoblot analysis of Col and rif1 seedling extracts with
specific antibodies against the plastome-encoded proteins
AtpB (for ATPase b chain) and PsbA (for photosystem II protein
D1) confirmed a decreased production of plastome-encoded
proteins in the absence of RIF1 function (Figure 7). As expected,
which is defective in plastidial RNA processing (Sauret-Gu ¨eto
et al., 2006), and in Col seedlings grown in the presence of
sublethal concentrations of chloramphenicol (CAP), an inhibitor
of plastid protein synthesis. In all cases, partially impaired
synthesis of plastome-encoded proteins correlated with con-
comitantly upregulated DXR levels (Figure 7), confirming our
previous observations (Sauret-Gu ¨eto et al., 2006).
Homeostasis of DXS and DXR Levels Is Controlled by the
Plastidial Clp Protease Complex
The discovery that the expression of the plastid genome is
impaired in the rif1 mutant and causes similar phenotypes to
those described for the rif10 mutant and CAP-treated wild-type
seedlings (Sauret-Gu ¨eto et al., 2006) confirms a link between
plastome-encoded proteins and the regulation of MEP pathway
enzyme levels in plastids. Most of the ;100 genes found in the
plastid genome encode proteins of the photosynthetic appa-
ratus and the transcription/translation system. The only plas-
tome genes of known function in higher plants that fall out of
ClpP1, a catalytic component of the plastidic ATP-dependent
Clp protease (Wakasugi et al., 2001). Since the Clp protease
has been proposed to target nucleus-encoded chloroplast-
imported proteins (Sjo ¨gren et al., 2006) and decreased
Figure 5. Putative Role of NO in the Regulation of the MEP Pathway.
(A) Representative photographs of Col and rif1-1 seedlings germinated
and grown for 5 d on MS plates containing a NO donor (50 mM SNP) or a
NO donor analog as a control (50 mM SFC) and either supplemented (þ)
or not (?) with 50 mM FSM.
(B) Immunoblot analysis of DXR levels in samples grown without FSM as
described for (A). The position of the DXR protein is indicated with an
arrowhead. The unspecific band recognized by the antibody is also
shown as a loading control.
1308The Plant Cell
proteolytic degradation might be a plausible explanation for the
increased levels of MEP pathway enzymes in rif1 plastids, we
aimed to analyze the role of ClpP1 and the plastidic Clp pro-
tease complex in the posttranscriptional regulation of the MEP
Unexpectedly, the levels of plastome-encoded ClpP1 protein
were not reduced but increased in mutant rif1 and rif10 and in
CAP-treated Col seedlings compared with untreated wild-type
controls (Figure 7). ClpP1 together with four more Ser-type
proteases (ClpP3 to ClpP6) and four related nonproteolytic
proteins (ClpR1 to ClpR4) form the catalytic core of the Clp
complex. Because the formation of a functionally active Clp
protease relies on the correct stoichiometry of each of these
subunits (Sjo ¨gren et al., 2006), the observed upregulation of
ClpP1 in response to an impaired expression of the plastome
might be a compensatory response to the inability to form
sufficient active stromal Clp protease (Sjo ¨gren et al., 2004). To
confirm whether a reduction in the activity of the plastidial Clp
proteolytic complex had an effect on MEP pathway enzyme
mutant (SALK_088407), in which the loss of function of nucleus-
encoded ClpR1 causes a reduction of other ClpPR subunits,
including ClpP1(Koussevitzkyetal.,2007).As showninFigure8,
Figure 6. Ultrastructure of Plastids in Wild-Type and rif1-1 Cotyledons.
Seedlings were germinated and grown on MS plates in the dark (etiolated) or under long-day conditions for 3 d, and cotyledons were then examined by
transmission electron microscopy. Mitochondria are indicated with arrowheads. Bars ¼ 0.5 mm in (A) to (C) and 1 mm in (D) and (E).
(A) Etiolated Col seedling. The prolamellar body is clearly visible in the central-right area of the etioplast.
(B) and (C) Etiolated rif1 seedling. Plastids lack a prolamellar body and contain electrodense plastoglobuli-like vesicles.
(D) Light-grown Col seedling.
(E) Light-grown rif1 seedling.
MEP Pathway Regulation by Clp Protease1309
a posttranscriptional increase in the levels of DXS and DXR
compared with wild-type siblings, supporting a role of the
plastidic Clp protease complex in the degradation of these
MEP pathway enzymes. Because stromal substrates of the
plastidic Clp machinery are reportedly more abundant in devel-
oping chloroplasts of young expanding leaves (Sjo ¨gren et al.,
2006), we next compared the levels of DXS and DXR proteins in
plants with those in fully expanded leaves of the outer whorl.
Immunoblot analyses showed that DXS and DXR were much
more abundant in younger leaves, whereas no differences were
observed at the transcript level (Figure 8), providing additional
evidence that the stromal Clp protease could be involved in the
degradation of MEP pathway enzymes in functional wild-type
To further investigate whether DXS and DXR could be sub-
strates for the plastidic Clp protease complex, we performed an
in vivo degradation assay as described (Sjo ¨gren et al., 2006) and
compared the stability of these stromal proteins in wild-type and
mutant rif1 chloroplasts. Intact chloroplasts were incubated in
the light with ATP, and samples were taken at different time
points for immunoblot analysis. Figure 9 shows that both DXS
and DXR proteins were degraded at a lower rate in mutant rif1
chloroplasts. As a result, the ratio of DXS levels in mutant versus
wild-type chloroplasts was increased by 60% after 1 h of incu-
bation in degradation buffer (from 1.5 to 2.4). A similar increase
(65%, from 2.0 to 3.3) was observed for DXR (Figure 9). Our
results are in agreement with these two MEP pathway enzymes
being targets of the stromal Clp protease complex.
A New Function for the RIF1 Protein in Chloroplasts
The RIF1/NOS1/NOA1 protein was first described to function as
a NO synthase (Guo et al., 2003), but concerns were later raised
about this activity (Crawford et al., 2006; Zemojtel et al., 2006a).
Although our data do not rule out the participation of RIF1 in
the production of NO in Arabidopsis, the fact that not all of the
phenotypes of rif1 plants can be rescued by treatment with the
NO donor SNP (Figure 5) suggests that the loss of RIF1 function
affects other processes unrelated to NO synthesis. In particular,
the increase in DXR levels eventually responsible for the FSM
resistance phenotype of mutant rif1 seedlings was unaffected
by treatment with SNP (Figure 5), indicating that the posttran-
scriptional accumulation of MEP pathway enzymes is not reg-
ulated by NO levels but most likely by an alternative function of
RIF1. Although it has been reported that this protein is targeted
to mitochondria in root cells (Guo and Crawford, 2005), here we
provide several lines of evidence that RIF1 functions predom-
inantly in plastids. RIF1 was imported into wild-type chloro-
plasts in vitro (Figure 4D), and a functional RIF1-GFP fusion
protein was found to be localized in chloroplasts of rescued rif1
chloroplasts but not mitochondria was clearly affected in rif1
seedlings (Figure 6) further supports a function of RIF1 in differ-
ent types of plastids.
RIF1 is most similar to GTP binding proteins of the YlqF/
YawG family, which are found in many prokaryotic and eukary-
otic organisms (Leipe et al., 2002; Zemojtel et al., 2006a).
Increasing evidence shows that prokaryotic members of this
family are involved in the biogenesis, assembly, and/or stability
of bacterial ribosomes by acting as chaperones or enzymes
modifying rRNA or protein functions (Comartin and Brown,
2006; Uicker et al., 2007). The closest RIF1 homolog is YqeH
(Figure 4C), a B. subtilis protein required for the formation of the
bacterial 70S ribosome and the assembly or stability of the
small (30S) ribosomal subunit (Uicker et al., 2007). Homologs
from eukaryotic organisms lacking plastids, such as yeast and
mammals, have been experimentally found in mitochondria,
and the yeast YqeH protein (YOR205C) was shown to copurify
in a complex with mitochondrial ribosomal proteins of the small
subunit and to interact with mitochondrial translation elonga-
tion factor EF1a (Zemojtel et al., 2006b). Consistent with RIF1
having a similar activity in plastids, the mature protein contains
an N-terminal Zn ribbon domain (see Supplemental Figure 2
Figure 7. Levels of Plastidial Proteins in Seedlings with Impaired Ex-
pression of the Plastome.
Proteins were extracted from Col, rif1-1, and rif10-1 seedlings grown
under long-day conditions for 5 d on MS plates (left panels) and from Col
seedlings grown under the same conditions on MS plates supplemented
(þ) or not (?) with 15 mM CAP (right panels). Immunoblot analyses were
performed with antibodies raised against the indicated proteins. The
position of the DXR protein is indicated with an arrowhead. The arrow
marks the position of the RBCL protein detected by Coomassie blue
1310 The Plant Cell
online) that might be involved in binding to RNA (Zemojtel et al.,
2006a). Also, a recombinant version of the bacterial YqeH
protein targeted to plastids of rif1 plants was able to comple-
ment the mutant phenotype (Figure 4B; see Supplemental
Figure 3 online). These results, together with the observed
decrease in the levels of plastome-encoded proteins, support a
main role for RIF1 in the expression of the plastid genome
(plastome), most likely in a process required for normal plastid
ribosome function. This conclusion is further supported by the
striking phenotypic similarities between rif1 and rif10, a mutant
known to be defective in the processing of all types of plastid
RNA, including rRNA (Sauret-Gu ¨eto et al., 2006).
The Levels of Key MEP Pathway Enzymes Are Modulated
by the Plastidic Clp Protease
The results wehave shownhere and in aprevious report (Sauret-
Gu ¨eto et al., 2006) confirm a strong and specific influence of
plastid cues in the regulation of the MEP pathway for isoprenoid
biosynthesis. Evidence provided in this work unveils a mecha-
nism for such regulation involving the participation of the
plastidic Clp protease complex. Impaired expression of the
plastid genome in rif1, rif10, and CAP-treated Col seedlings
unexpectedly resulted in increased levels of the plastome-
encoded ClpP1 subunit of the catalytic ClpPR core of the
complex (Figure 7). It is possible that ClpP1 levels are modulated
not only by their biosynthetic rate but also by a regulatory
feedback mechanism at the posttranslational level. As a result, a
defective production of ClpP1 in the first stages of plastid
development might result in an altered proportion of subunits
within the ClpPR core and an insufficient Clp protease activity,
which in turn might lead to the observed upregulation of ClpP1
importantly, altered levels of ClpPR subunits in mutant clpr1-2
seedlings (Koussevitzky et al., 2007) also result in enhanced
levels of both DXS and DXR and a concomitant phenotype of
FSM resistance (Figure 8). The fact that these MEP pathway
wild-type plants compared with fully expanded leaves of the
outer whorl of the rosette (Figure 8), as expected for most
substrates of the plastidic Clp complex (Sjo ¨gren et al., 2006),
Figure 8. Role of the Plastidial Clp Protease Complex in the Degradation
of DXS and DXR.
(A) Representative photographs of Col and mutant clpr1-2 seedlings
germinated and grown under long-day conditions for 5 d on MS plates
supplemented (þ) or not (?) with 50 mM FSM.
(B) Immunoblot analysis of DXS and DXR levels in protein extracts from
seedlings grown as described for (A) without FSM (left panels) and from
leaves of the inner (I) or outer (O) whorls of the rosette of 3-week-old Col
plants (right panels). The position of the DXR protein is indicated with an
arrowhead. The arrow marks the position of the RBCL protein detected
by Coomassie blue (C-Blue) staining.
(C) Real-time quantitative RT-PCR analysis of transcript levels of the
indicated genes in the seedling and rosette leaf samples described for
(B). Levels are represented relative to those in Col seedlings and
correspond to the mean and SE from duplicate PCR analyses of two
MEP Pathway Regulation by Clp Protease1311
further implicates DXS and DXR as targets of the plastidial Clp
protease. Consistently, a lower degradation rate of these two
stromal enzymes was observed in rif1 compared with wild-type
chloroplasts (Figure 9).
Proteolytic activities are required for the biogenesis and func-
tioning of plastids and for their adaptation to changing environ-
mental conditions. Plastid proteases of different classes remove
and degrade targeting peptides, eliminate and recycle dam-
aged, misfolded, or misassembled proteins and complexes,
and help maintain the correct stoichiometry between different
plastid Clp complex is the most prominent stromal protease,
and it shows far more complexity in higher plants than in any
other organism. The presence of similar Clp core complexes in
all plastid types, their abundance, and the essential character
of the catalytic subunits has led to the proposal that this prote-
ase plays a central role in plastid homeostasis and biogenesis,
similar to that of the ubiquitin-proteasome system for plant
development(Peltieretal.,2004).Incontrast withthe impressive
progress in the structural characterization of this complex, lit-
tle is currently known concerning the specific function of this
stromal protease (Adam et al., 2006; Sakamoto, 2006). Genetic
approaches have revealed that ClpP1 and other subunits of the
catalytic ClpPR core are required for the correct differentiation
of plastids and other aspects of plant development (Shikanai
et al., 2001; Kuroda and Maliga, 2003; Rudella et al., 2006;
Sjo ¨gren et al., 2006; Zheng et al., 2006; Koussevitzky et al.,
2007), but the identification of protein substrates in higher plants
has remained elusive. Several putative substrates, including a
variety of chaperones and components required for the synthe-
sis of plastome-encoded proteins, were recently identified
by searching for stromal proteins differentially accumulated in
chloroplasts from transgenic plants with reduced levels of Clp
protease activity (Sjo ¨gren et al., 2004, 2006; Rudella et al.,
2006). Our results (Figures 8 and 9) support the hypothesis that
some of the MEP pathway enzymes, which are also localized in
the stroma (Carretero-Paulet et al., 2002; Oudin et al., 2007),
could be targets of the plastidic Clp protease machinery as well.
However, current evidence does not allow us to conclude
whether these proteins are direct targets (actual substrates) of
this protease. It is possible that downregulation of Clp activity in
plastids results in changes that are ultimately responsible for the
observed accumulation of DXS, DXR, and/orthe other proposed
substrates. For example, the observed upregulation of a variety
of stromal chaperones in plants with reduced levels of Clp pro-
tease (Sjo ¨gren et al., 2004, 2006; Rudella et al., 2006) might
contribute to the observed phenotype of higher DXS and DXR
levels and FSM resistance.
Although more experiments are needed to fully establish how
the plastidic Clpprotease and/or other plastid proteins modulate
an important step forward to understanding the regulation of
isoprenoid production in plastids. Changes in light conditions
and photosynthetic activity can affect the synthesis of plastome-
but also the accumulation of plastidic Clp complex subunits
(Zheng et al., 2006), which in turn could result in changes in the
levels of DXS and DXR proteins. Because both DXS and DXR
have been shown to control flux through the MEP pathway in
Arabidopsis (Este ´vez et al., 2001; Carretero-Paulet et al., 2006)
and other plants (Rodrı ´guez-Concepcio ´n, 2006), the described
posttranscriptional mechanism could allow individual plastids to
rapidly optimize the supply of isoprenoid precursors when
needed for the biosynthesis of end products.
Figure 9. Degradation of DXS and DXR in Isolated Chloroplasts.
(A) Chloroplasts isolated from Col and rif1-1 seedlings grown for 2 weeks on MS plates under long-day conditions were incubated for 1 h in the
light at 258C in the presence of 5 mM ATP. Aliquots were taken at the indicated times (minutes) and used for protein extraction and immunoblot
analysis with antibodies against DXS and DXR. The amount of protein loaded was doubled in the case of Col samples, so bands of similar
intensity were obtained in lanes corresponding to untreated Col and rif1 chloroplasts (time 0) in order to better compare degradation rates. The
position of the DXR protein is indicated with an arrowhead. The other major band recognized by the anti-DXR serum is shown to monitor protein
(B) Quantification of protein levels from two degradation experiments performed as described for (A). DXS and DXR levels (from immunoblot band
intensity) were normalized to those of the unspecific band recognized by the anti-DXR serum and are represented relative to the level in untreated Col
chloroplasts. Mean and SE from duplicate blots of two experiments are represented. Boldface numbers above the columns indicate the ratio between
DXS or DXR protein levels in rif1 versus Col chloroplasts.
1312The Plant Cell
Plant Material and Growth Conditions
Arabidopsis thaliana seeds from activation-tagging T-DNA collections
(Weigel et al., 2000), Salk T-DNA insertion lines (Alonso et al., 2003), and
the Ds-GeneTrap line GT6235 (Martienssen, 1998) were obtained from
the Nottingham Arabidopsis Stock Centre. Wild-type and mutant geno-
types used in this work were in the Col background with the exception of
the GT6235 line, which was generated in Landsberg erecta. Seeds were
surface-sterilized before plating on Petri dishes with solid Murashige and
Skoog (MS) medium as described (Sauret-Gu ¨eto et al., 2006). Where
indicated, the medium was supplemented with FSM (Gateway Chemical
Technology), SNP (Sigma-Aldrich), SFC (Sigma-Aldrich), or CAP (Sigma-
Aldrich). Plates were kept for at least 2 d at 48C for stratification and then
transferredtoagrowthchamber at228C underlong-dayconditions(8hin
the dark and 16 h under fluorescent white light at a photon fluence rate of
plates to a 1:1:1 (v/v) perlite:vermiculite:sphagnum soil mixture irrigated
with mineral nutrients and grown in the long-day chamber.
Identification of the T-DNA Insertion Site and Sequence Analysis
For the identification of the gene responsible for the FSM resistance
phenotype in rif1, homozygous mutant plants were backcrossed with the
Col wild type to test whether the corresponding mutations were linked to
the presence of the T-DNA. After identifying the recessive nature of the
rif1 mutation and its linkage to the resistance marker associated with the
T-DNA, the insertion site was identified using an inverse PCR strategy as
described (Sauret-Gu ¨eto et al., 2006).
Sequence analyses were performed with the Vector NTI Suite 6
(InforMax) software package and the available web resources. RIF1
putative homologs were searched on the National Center for Biotechnol-
ogy Information and The Arabidopsis Information Resource databases
(www.ncbi.nlm.nih.gov and www.Arabidopsis.org,respectively). Multiple
alignments were performed with the ClustalW program (www2.ebi.ac.uk/
clustalw), edited with GeneDoc (www.psc.edu/biomed/genedoc), and
used to construct a phylogenetic tree with MEGA 2.1 (Molecular Evolu-
tionary Genetics Analysis; www.megasoftware.net).
Generation of Transgenic Plants
Constructs for the constitutive expression of the fusion proteins RIF1-
GFP and P-YqeH-GFP in transgenic plants were made as described in the
Supplemental Methods online and used for Agrobacterium tumefaciens–
mediated transformation of rif1-1 plants as described (Sauret-Gu ¨eto
et al., 2006). Plants expected to harbor the transgenes were selected
based on their ability to survive when grown on MS plates supplemented
with 50 mg/mL hygromycin. The resulting positive plants were transferred
to soil and allowed to set seed. Lines harboring a single T-DNA insertion
were identified by segregation of the hygromycin selection marker and
used for analysis.
Analysis of Transcript and Protein Levels
Real-time quantitative RT-PCR was performed as described (Carretero-
Paulet et al., 2006). Crude total protein extracts from Arabidopsis tissues
were obtained by grinding samples in liquid nitrogen and resuspending in
ice-cold TKMEShomogenization buffer(100mMTricine-KOH, pH7.5,10
mM KCl, 1 mM MgCl2, 1 mM EDTA, and 10% [w/v] sucrose) supple-
mented with 0.2% (v/v) Triton X-100, 1 mM DTT, 100 mg/mL phenyl-
methylsulfonyl fluoride, 3 mg/mL E64, and 20 mL/mL protease inhibitor
cocktail (Sigma-Aldrich). Protein concentration was determined as de-
scribed (Carretero-Paulet et al., 2002). Separation by SDS-PAGE
and immunoblot analyses were performed as described (Rodrı ´guez-
Concepcio ´n et al., 2004). Antibodies against DXS, DXR, and ClpP1 were
kind gifts of Patricia Leo ´n (Instituto de Biotecnologı ´a-UNAM), Michael H.
Walter (Leibniz Institute of Plant Biochemistry), and Zach Adam (Hebrew
University), respectively. Sera against AtpB and PsbA were purchased
decreased levels of RBCL (the major protein in the extracts), protein
loading was normalized to the levels of other proteins detected by
Coomassie Brilliant Blue staining of the gels. Bands corresponding to
DXS and DXR were identified by size and by comparison with samples
from transgenic overexpression lines (Carretero-Paulet et al., 2006;
Sauret-Gu ¨eto et al., 2006). Chemiluminescent signals of the selected
bands were visualized using a LAS-3000 (Fujifilm) image analyzer and
quantified with the Multigauge Fujifilm 3.0 software. The levels of the
unspecific band recognized by the anti-DXR serum were also quantified
for normalization and used as an additional control of equal loading.
Similar results were obtained when different proteins detected by
Coomassie Brilliant Blue staining were used for normalization.
For chloroplast import experiments, a cDNA sequence encoding the full-
length RIF1 protein was cloned into the SmaI site of pBluescript SKþ
(Stratagene). A construct with the right orientation, pBS-RIF1, was used
as atemplatefor in vitro transcription and translation using theTNT Quick
T7 system (Promega) with35S-labeled Met and T7 RNA polymerase.
Plasmid pDXR-At (Carretero-Paulet et al., 2002) was used to obtain
labeled DXR protein as a positive control. Import experiments were
performed with chloroplasts isolated from 10-d-old Arabidopsis seed-
lings as described (Kubis et al., 2007) (see also Supplemental Methods
online). Import was performed in white light at 258C for 10 min. Thermo-
lysin treatment and detection of imported polypeptides were performed
as described (Aronsson and Jarvis, 2002).
For degradation experiments, chloroplasts isolated from 14-d-old
seedlings were resuspended in HMS buffer (see Supplemental Methods
online) supplemented with 20 mM gluconic acid, 10 mM NaHCO3, 0.2%
(w/v) BSA, and 5 mM Mg-ATP and incubated at 258C under white light
(100 mmol?m?2?s?1). Aliquots of ;107chloroplasts were taken at 0, 15,
30, and 60 min, immediately pelleted, and frozen in liquid nitrogen to stop
all protein degradation, as described (Sjo ¨gren et al., 2004). Pelleted
samples were resuspended in ice-cold TKMES buffer (see Supplemental
Methods online) and incubated for 15 min on ice for chloroplast lysis and
protein solubilization. Determination of protein concentration, separation
by SDS-PAGE, and immunoblot analysis were performed as described
Methods for the observation and recording of chlorophyll autofluores-
cence in whole seedlings, confocal laser scanning microscopy, and
transmission electron microscopy are described in the Supplemental
Sequence data from this article can be found in the GenBank/EMBL data
libraries under the following accession numbers: AtpB, AtCg00480;
At4g15560; HDR, At4g34350; HDS, At5g60600; MJ1464, Q58859;
PsbA, AtCg00020; RBCL, AtCg00490; RIF1/NOS1/NOA1, At3g47450;
RIF10, At3g03710; YawG, Q10190; YjeQ, P39286; YlqF, F69880; and
MEP Pathway Regulation by Clp Protease 1313
The following materials are available in the online version of this article.
Supplemental Figure 1. Quantification of FSM Resistance of rif1
Supplemental Figure 2. Multiple Alignment of the B. subtilis YqeH
Protein with the Arabidopsis Closest Homologs.
Supplemental Figure 3. Phenotype of rif1 Plants Constitutively
Expressing a Plastid-Targeted Bacterial YqeH Protein Fused to GFP
Supplemental Figure 4. Ultrastructure of Chloroplasts in Wild-Type
and rif1 Leaves.
Supplemental Methods. Constructs for the Production of Transgenic
Lines; Chloroplast Isolation; Microscopy.
Supplemental Data Set 1. Text File of the Alignment Corresponding
to Figure 4C and Supplemental Figure 2 online.
We thank J. Martı ´nez-Garcı ´a and M.A. Phillips for critical reading of the
manuscript, S. Kubis and J. Be ´dard for help and advice with the import
experiments, and P. Leo ´n, M. Walter, and Z. Adam for the gift of anti-
bodies. We are also grateful to the Nottingham Arabidopsis Stock Centre,
the Salk Institute Genomic Analysis Laboratory, and the Cold Spring
Harbor Laboratory for valuable seed and information resources. The
excellent technical support from A. Orozco and the staffs of the Serveis
Cientificote `cnics and the Serveis de Camps Experimentals of the Uni-
versitat de Barcelona and the Greenhouse and Microscopy facilities at the
Centre for Research on Agricultural Genomics is greatly appreciated. This
work was supported by grants from the Generalitat de Catalunya
(Distincio ´) and the Spanish Ministerio de Ciencia y Tecnologı ´a and FEDER
to M.R.-C. (Grant BIO2005-00357). U.F.-P. and S.S.-G. received PhD
fellowships from the Spanish Ministerio de Educacio ´n y Ciencia (FPU
program) and the Generalitat de Catalunya, respectively.
Received February 11, 2008; revised March 18, 2008; accepted April 22,
2008; published May 9, 2008.
Adam, Z., Rudella, A., and van Wijk, K.J. (2006). Recent advances in
the study of Clp, FtsH and other proteases located in chloroplasts.
Curr. Opin. Plant Biol. 9: 234–240.
Alonso, J.M., et al. (2003). Genome-wide insertional mutagenesis of
Arabidopsis thaliana. Science 301: 653–657.
Aronsson, H., and Jarvis, P. (2002). A simple method for isolating import-
competent Arabidopsis chloroplasts. FEBS Lett. 529: 215–220.
Botella-Pavı ´a, P., Besumbes, O., Phillips, M.A., Carretero-Paulet, L.,
Boronat, A., and Rodrı ´guez-Concepcio ´n, M. (2004). Regulation of
carotenoid biosynthesis in plants: Evidence for a key role of hydroxy-
methylbutenyl diphosphate reductase in controlling the supply of
plastidial isoprenoid precursors. Plant J. 40: 188–199.
Budziszewski, G.J., et al. (2001). Arabidopsis genes essential for
seedling viability. Isolation of insertional mutants and molecular clon-
ing. Genetics 159: 1765–1778.
Carretero-Paulet, L., Ahumada, I., Cunillera, N., Rodrı ´guez-Concepcio ´n,
M., Ferrer, A., Boronat, A., and Campos, N. (2002). Expression and
molecular analysis of the Arabidopsis DXR gene encoding 1-deoxy-
D-xylulose 5-phosphate reductoisomerase, the first committed enzyme
of the 2-C-methyl-D-erythritol 4-phosphate pathway. Plant Physiol. 129:
Carretero-Paulet, L., Cairo, A., Botella-Pavia, P., Besumbes, O.,
Campos, N., Boronat, A., and Rodriguez-Concepcion, M. (2006).
Enhanced flux through the methylerythritol 4-phosphate pathway in
Arabidopsis plants overexpressing deoxyxylulose 5-phosphate re-
ductoisomerase. Plant Mol. Biol. 62: 683–695.
Comartin, D.J., and Brown, E.D. (2006). Non-ribosomal factors in
ribosome subunit assembly are emerging targets for new antibacterial
drugs. Curr. Opin. Pharmacol. 6: 453–458.
Crawford, N.M., Gally, M., Tischner, R., Heimer, Y.M., Okamoto, M.,
and Mack, A. (2006). Plant nitric oxide synthase: Back to square one.
Trends Plant Sci. 11: 526–527.
Eisenreich, W., Bacher, A., Arigoni, D., and Rohdich, F. (2004).
Biosynthesis of isoprenoids via the non-mevalonate pathway. Cell.
Mol. Life Sci. 61: 1401–1426.
Enfissi, E.M.A., Fraser, P.D., Lois, L.M., Boronat, A., Schuch, W., and
Bramley, P.M. (2005). Metabolic engineering of the mevalonate and
non-mevalonate isopentenyldiphosphate-forming pathways for the pro-
duction of health-promoting isoprenoids in tomato. Plant Biotechnol. J.
Este ´vez, J.M., Cantero, A., Reindl, A., Reichler, S., and Leo ´n, P.
(2001). 1-Deoxy-D-xylulose-5-phosphate synthase, a limiting enzyme
for plastidic isoprenoid biosynthesis in plants. J. Biol. Chem. 276:
Guevara-Garcia, A., San Roman, C., Arroyo, A., Cortes, M.E.,
Gutierrez-Nava, M.L., and Leon, P. (2005). Characterization of the
Arabidopsis clb6 mutant illustrates the importance of posttranscrip-
tional regulation of the methyl-D-erythritol 4-phosphate pathway.
Plant Cell 17: 628–643.
Guo, F.Q., and Crawford, N.M. (2005). Arabidopsis nitric oxide syn-
thase1 is targeted to mitochondria and protects against oxidative
damage and dark-induced senescence. Plant Cell 17: 3436–3450.
Guo, F.Q., Okamoto, M., and Crawford, N.M. (2003). Identification of a
plant nitric oxide synthase gene involved in hormonal signaling.
Science 302: 100–103.
Gutierrez-Nava, M.L., Gillmor, C.S., Jimenez, L.F., Guevara-Garcia,
A., and Leon, P. (2004). CHLOROPLAST BIOGENESIS genes act cell
and noncell autonomously in early chloroplast development. Plant
Physiol. 135: 471–482.
Kasahara, H., Hanada, A., Kuzuyama, T., Takagi, M., Kamiya, Y., and
Yamaguchi, S. (2002). Contribution of the mevalonate and methyl-
erythritol phosphate pathways to the biosynthesis of gibberellins in
Arabidopsis. J. Biol. Chem. 277: 45188–45194.
Koussevitzky, S., Stanne, T.M., Peto, C.A., Giap, T., Sjogren, L.L.,
Zhao, Y., Clarke, A.K., and Chory, J. (2007). An Arabidopsis thaliana
virescent mutant reveals a role for ClpR1 in plastid development. Plant
Mol. Biol. 63: 85–96.
Kubis, S.E., Lilley, K.S., and Jarvis, P. (2007). Isolation and preparation
of chloroplast from Arabidopsis thaliana plants. In Methods in Molec-
ular Biology, A. Posch, ed (Totowa, NJ: Humana Press), pp. 171–186.
Kuroda, H., and Maliga, P. (2003). The plastid clpP1 protease gene is
essential for plant development. Nature 425: 86–89.
Laule, O., Furholz, A., Chang, H.S., Zhu, T., Wang, X., Heifetz, P.B.,
Gruissem, W., and Lange, M. (2003). Crosstalk between cytosolic
and plastidial pathways of isoprenoid biosynthesis in Arabidopsis
thaliana. Proc. Natl. Acad. Sci. USA 100: 6866–6871.
Leipe, D.D., Wolf, Y.I., Koonin, E.V., and Aravind, L. (2002). Classi-
fication and evolution of P-loop GTPases and related ATPases.
J. Mol. Biol. 317: 41–72.
Lichtenthaler, H.K. (1999). The 1-deoxy-D-xylulose-5-phosphate path-
way of isoprenoid biosynthesis in plants. Annu. Rev. Plant Physiol.
Plant Mol. Biol. 50: 47–65.
1314 The Plant Cell
Lo ´pez-Juez, E. (2007). Plastid biogenesis, between light and shadows. Download full-text
J. Exp. Bot. 58: 11–26.
Mahmoud, S.S., and Croteau, R.B. (2001). Metabolic engineering of
essential oil yield and composition in mint by altering expression of
deoxyxylulose phosphate reductoisomerase and menthofuran syn-
thase. Proc. Natl. Acad. Sci. USA 98: 8915–8920.
Mandel, M.A., Feldmann, K.A., Herrera-Estrella, L., Rocha-Sosa, M.,
and Leon, P. (1996). CLA1, a novel gene required for chloroplast
development, is highly conserved in evolution. Plant J. 9: 649–658.
Martienssen, R. (1998). Functional genomics: Probing plant gene
function and expression with transposons. Proc. Natl. Acad. Sci.
USA 95: 2021–2026.
Nagata, N., Suzuki, M., Yoshida, S., and Muranaka, T. (2002).
Mevalonic acid partially restores chloroplast and etioplast develop-
ment in Arabidopsis lacking the non-mevalonate pathway. Planta 216:
Oudin, A., Mahroug, S., Courdavault, V., Hervouet, N., Zelwer, C.,
Rodriguez-Concepcion, M., St-Pierre, B., and Burlat, V. (2007).
Spatial distribution and hormonal regulation of gene products from
methyl erythritol phosphate and monoterpene-secoiridoid pathways
in Catharanthus roseus. Plant Mol. Biol. 65: 13–30.
Peltier, J.B., Ripoll, D.R., Friso, G., Rudella, A., Cai, Y., Ytterberg, J.,
Giacomelli, L., Pillardy, J., and van Wijk, K.J. (2004). Clp protease
complexes from photosynthetic and non-photosynthetic plastids and
mitochondria of plants, their predicted three-dimensional structures,
and functional implications. J. Biol. Chem. 279: 4768–4781.
Pfannschmidt, T. (2002). Chloroplast redox signals: How photosynthe-
sis controls its own genes. Trends Plant Sci. 8: 33–41.
Querol, J., Campos, N., Imperial, S., Boronat, A., and Rodriguez-
Concepcion, M. (2002). Functional analysis of the Arabidopsis
thaliana GCPE protein involved in plastid isoprenoid biosynthesis.
FEBS Lett. 514: 343–346.
Rodrı ´guez-Concepcio ´n, M. (2006). Early steps in isoprenoid biosyn-
thesis: Multilevel regulation of the supply of common precursors in
plant cells. Phytochem. Rev. 5: 1–15.
Rodrı ´guez-Concepcio ´n, M., and Boronat, A. (2002). Elucidation of the
methylerythritol phosphate pathway for isoprenoid biosynthesis in
bacteria and plastids. A metabolic milestone achieved through ge-
nomics. Plant Physiol. 130: 1079–1089.
Rodrı ´guez-Concepcio ´n,M., Fore ´s,
Gonza ´lez, V., Phillips, M.A., Ferrer, A., and Boronat, A. (2004).
Distinct light-mediated pathways regulate the biosynthesis and ex-
change of isoprenoid precursors during Arabidopsis seedling devel-
opment. Plant Cell 16: 144–156.
O.,Martı ´nez-Garcı ´a,J.F.,
Rudella, A., Friso, G., Alonso, J.M., Ecker, J.R., and van Wijk, K.J.
(2006). Downregulation of ClpR2 leads to reduced accumulation of
the ClpPRS protease complex and defects in chloroplast biogenesis
in Arabidopsis. Plant Cell 18: 1704–1721.
Sakamoto, W. (2006). Protein degradation machineries in plastids.
Annu. Rev. Plant Biol. 57: 599–621.
Sauret-Gu ¨eto, S., Botella-Pavia, P., Flores-Perez, U., Martinez-
Garcia, J.F., San Roman, C., Leon, P., Boronat, A., and Rodriguez-
Concepcion, M. (2006). Plastid cues posttranscriptionally regulate the
accumulation of key enzymes of the methylerythritol phosphate
pathway in Arabidopsis. Plant Physiol. 141: 75–84.
Shikanai, T., Shimizu, K., Ueda, K., Nishimura, Y., Kuroiwa, T., and
Hashimoto, T. (2001). The chloroplast clpP gene, encoding a proteo-
lytic subunit of ATP-dependent protease, is indispensable for chloro-
plast development in tobacco. Plant Cell Physiol. 42: 264–273.
Sjo ¨gren, L.L., MacDonald, T.M., Sutinen, S., and Clarke, A.K. (2004).
Inactivation of the clpC1 gene encoding a chloroplast Hsp100 mo-
lecular chaperone causes growth retardation, leaf chlorosis, lower
photosynthetic activity, and a specific reduction in photosystem
content. Plant Physiol. 136: 4114–4126.
Sjo ¨gren, L.L., Stanne, T.M., Zheng, B., Sutinen, S., and Clarke, A.K.
(2006). Structural and functional insights into the chloroplast ATP-
dependent Clp protease in Arabidopsis. Plant Cell 18: 2635–2649.
Uicker, W.C., Schaefer, L., Koenigsknecht, M., and Britton, R.A.
(2007). The essential GTPase YqeH is required for proper ribosome
assembly in Bacillus subtilis. J. Bacteriol. 189: 2926–2929.
Wakasugi, T., Tsudzuki, T., and Sugiura, M. (2001). The genomics of
land plant chloroplasts: Gene content and alteration of genomic
information by RNA editing. Photosynth. Res. 70: 107–118.
Weigel, D., et al. (2000). Activation tagging in Arabidopsis. Plant
Physiol. 122: 1003–1013.
Zemojtel, T., Frohlich, A., Palmieri, M.C., Kolanczyk, M., Mikula, I.,
Wyrwicz, L.S., Wanker, E.E., Mundlos, S., Vingron, M., Martasek,
P., and Durner, J. (2006a). Plant nitric oxide synthase: A never-
ending story? Trends Plant Sci. 11: 524–525.
Zemojtel, T., Kolanczyk, M., Kossler, N., Stricker, S., Lurz, R., Mikula,
I., Duchniewicz, M., Schuelke, M., Ghafourifar, P., Martasek, P.,
Vingron, M., and Mundlos, S. (2006b). Mammalian mitochondrial nitric
oxide synthase: Characterization of a novel candidate. FEBS Lett. 580:
Zheng, B., MacDonald, T.M., Sutinen, S., Hurry, V., and Clarke, A.K.
(2006). A nuclear-encoded ClpP subunit of the chloroplast ATP-
dependent Clp protease is essential for early development in Arabi-
dopsis thaliana. Planta 224: 1103–1115.
MEP Pathway Regulation by Clp Protease1315