Complexity of Hsp90 in organelle targeting

Abstract

Heat shock protein 90 (Hsp90) is an abundant and highly conserved molecular chaperone. In Arabidopsis, the Hsp90 gene family consists of seven members. Here, we report that the AtHsp90-6 gene gives rise to two mRNA populations, termed AtHsp90-6L and AtHsp90-6S due to alternative initiation of transcription. The AtHsp90-6L and AtHsp90-6S transcription start sites are located 228 nucleotides upstream and 124 nucleotides downstream of the annotated translation start site, respectively. Both transcripts are detected under normal or heat-shock conditions. The inducibility of AtHsp90-6 mRNAs by heat shock implies a potential role of both isoforms in stress management. Stable transformation experiments with fusion constructs between the N-terminal part of each AtHsp90-6 isoform and green fluorescent protein indicated import of both fusion proteins into mitochondria. In planta investigation confirmed that fusion of the AtHsp90-5 N-terminus to green fluorescent protein (GFP) did result in specific chloroplastic localization. The mechanisms of regulation for mitochondria- and plastid-localized chaperone-encoding genes are not well understood. Future work is needed to address the possible roles of harsh environmental conditions and developmental processes on fine-tuning and compartmentalization of the AtHsp90-6L, AtHsp90-6S, and AtHsp90-5 proteins in Arabidopsis.

Full text

Complexity of Hsp90 in organelle targeting
Constantinos Prassinos Æ Kosmas Haralampidis Æ Dimitra Milioni Æ
Despina Samakovli Æ Konstantinos Krambis Æ Polydefkis Hatzopoulos
Received: 27 July 2007 / Accepted: 14 March 2008 / Published online: 27 March 2008
Ó Springer Science+Business Media B.V. 2008
Abstract Heat shock protein 90 (Hsp90) is an abundant
and highly conserved molecular chaperone. In Arabidopsis,
the Hsp90 gene family consists of seven members. Here,
we report that the AtHsp90-6 gene gives rise to two mRNA
populations, termed AtHsp90-6L and AtHsp90-6S due to
alternative initiation of transcription. The AtHsp90-6L and
AtHsp90-6S transcription start sites are located 228
nucleotides upstream and 124 nucleotides downstream of
the annotated translation start site, respectively. Both
transcripts are detected under normal or heat-shock con-
ditions. The inducibility of AtHsp90-6 mRNAs by heat
shock implies a potential role of both isoforms in stress
management. Stable transformation experiments with
fusion constructs between the N-terminal part of each
AtHsp90-6 isoform and green fluorescent protein indicated
import of both fusion proteins into mitochondria. In planta
investigation confirmed that fusion of the AtHsp90-
5 N-terminus to green fluorescent protein (GFP) did result
in specific chloroplastic localization. The mechanisms of
regulation for mitochondria- and plastid-localized chaper-
one-encoding genes are not well understood. Future work
is needed to address the possible roles of harsh environ-
mental conditions and developmental processes on fine-
tuning and compartmentalization of the AtHsp90-6L,
AtHsp90-6S, and AtHsp90-5 proteins in Arabidopsis.
Keywords Alternative transcription  Chloroplasts 
Heat shock  Hsp90  Mitochondria  Targeting
Introduction
Heat shock proteins 90 (Hsp90s) have been assigned
numerous and diverse functional roles in many pathways
ranging from cellular homeostasis and signal transduction
to development. Since they assist client protein functional
competence at a proper time and place, multiple cellular
compartments may require the presence of Hsp90.
The Hsp90s are evolutionarily conserved proteins found
from E. coli to humans (Emelyanov 2002). Under normal
conditions, Hsp90 function is essential for the biogenesis
and support of numerous cellular proteins that control cell
physiology. They have key roles in signal transduction,
protein folding, and protein degradation (Rutherford and
Lindquist 1998). Hsp90 proteins are molecular chaperones
acting to prevent misfolding and aggregation of unfolded
or partially folded proteins (Young et al. 2001). They assist
protein transport across the endoplasmic reticulum (ER)
and organellar membranes (Schatz and Dobberstein 1996).
In higher eukaryotes, Hsp90s form complexes with various
client proteins, including steroid hormone receptors (Joab
et al. 1984), helix-loop-helix transcription factors
C. Prassinos  D. Milioni  D. Samakovli  K. Krambis 
P. Hatzopoulos (&)
Laboratory of Molecular Biology, Agricultural Biotechnology
Department, Agricultural University of Athens, Iera Odos 75,
118 55 Athens, Greece
e-mail: phat@aua.gr
K. Haralampidis
Molecular Plant Development Laboratory, Faculty of Biology,
Department of Botany, University of Athens, 15784 Athens,
Greece
Present Address:
K. Krambis
Virginia Bioinformatics Institute, Washington Street - 0477,
Virginia Tech, Blacksburg, VA 24061, USA
123
Plant Mol Biol (2008) 67:323–334
DOI 10.1007/s11103-008-9322-8

(Wilhelmsson et al. 1990), tyrosine and serine/threonine
kinases, nitric oxide synthase and telomerase (Wegele
et al. 2004), the tumor suppressor p53 (Sepehrnia et al.
1996), and the type 1 tumor necrosis factor receptor
(TNFR-1) (Song et al. 1995). Recent studies in S. cerevi-
siae have revealed that Hsp90s interact to an extended
network of cofactors and substrates involved in a wide
range of cellular functions (Zhao et al. 2005).
Although different isoforms have been identified, a role
has not yet been assigned to each Hsp90 homolog, while
gene duplication may not result in functional redundancy.
The Arabidopsis genome contains seven Hsp90 protein-
encoding genes. These specify four cytoplasmic Hsp90
(AtHsp90-1, At5g52640; AtHsp90-2, At5g56030; AtHsp90-3,
At5g56010; AtHsp90-4, At5g56000), one ER Hsp90
(AtHsp90-7, At4g24190), and two organellar Hsp90 pro-
teins (AtHsp90-5, At2g04030; AtHsp90-6, At3g07770)
(Milioni and Hatzopoulos 1997; Krishna and Gloor 2001;
Chen et al. 2006). Genetic and biochemical studies are
only beginning to define their function and the nature of
their client proteins in plants. Studies in the fruit fly and
Arabidopsis strongly suggest that Hsp90 acts as a capacitor
for morphological evolution by most likely neutralizing the
conformational consequences of several mutations, there-
fore buffering their potential phenotypic changes and
turning them phenotypically silent (Rutherford and Lind-
quist 1998; Queitsch et al. 2002). Recent results in
Arabidopsis have shown that Hsp90 restricts stochastic
phenomena by minimizing perturbations, thereby canaliz-
ing development (Samakovli et al. 2007). The Arabidopsis
shd mutant phenotype could be the result of the accumu-
lation of cryptic mutations, promoting morphological
evolution in the Arabidopsis genome (Ishiguro et al. 2002).
In addition two Hsp90 isoforms, Hsp90-1 and Hsp90-2,
were identified as essential factors for the pathogen resis-
tance response mediated by several R proteins of the NB-
ARC-LRR class in plants (Takahashi et al. 2003; Zhang
et al. 2004; Sangster and Queitsch 2005; Boter et al. 2007).
Mitochondria and chloroplasts are the targets of differ-
ent stresses, often resulting in deleterious effects such as
apoptosis and cell cycle arrest. It has been suggested that
Hsp90 is engaged in chloroplast biogenesis during normal
growth and development (Cao et al. 2003). The CR88 gene
encodes a chloroplast-targeted Hsp90 homolog but its
crucial role remains to be elucidated (Cao et al. 2003). In
animals, a mitochondrial Hsp90 homolog (TRAP-1)
appears to have a crucial role in cell cycle progression,
cellular differentiation, and apoptosis (Felts et al. 2000;
Masuda et al. 2004).
Our results provide strong evidence for an exclusive
plastidic localization of AtHsp90-5 and demonstrate that
alternative transcription of the AtHsp90-6 gene generates
two different proteins transported into mitochondria.
Materials and methods
Plant material and transformation
Arabidopsis thaliana (ecotype Columbia) plants were
grown under standard conditions at 22°C under 70%
humidity with a light/dark cycle of 16 h/8 h. After trans-
formation, seeds from individual transgenic plants were
imbibed at 4°C overnight, surface sterilized, and layered on
MS medium containing 50 mg l
-1
kanamycin and
200 mg l
-1
cefotaxime, under the same growth conditions.
Transgenic plants were transferred to soil for further
analysis.
Arabidopsis thaliana protoplasts were isolated as pre-
viously described (Doelling and Pikaard 1993).
Primer extension analysis
Total RNA was extracted from 28-day-old Arabidopsis
plants using the RNeasy Plant Kit (Qiagen). Primer
extension analysis was carried out using the Primer
Extension System—AMV Reverse Transcriptase kit (Pro-
mega), according to the manufacturer’s instructions. The
following sequence-specific oligonucleotides were used:
PRET-1, 5
0
-ACGGAGCGCTTAGAGAGCCTGATC-3
0
(AtHsp90-6L) and PRET-2, 5
0
-CTGCACTGGATTCGTTC
CGGTATC-3
0
(AtHsp90-6S). PRET-1 and PRET-2 are
located 3 bp and 23 bp downstream of the translation start
sites, respectively.
RT-PCR analysis
Total RNA was extracted as described above. To assess heat
induction, 28-day-old plants were incubated at 37°C for 2 h.
After treatment, intact plants were harvested and used for
RNA preparation. First strand cDNA synthesis was carried
out using Expand Reverse Transcriptase (Roche Diagnos-
tics). The primers were as follows: for AtHsp90-6L, RC4 5
0
-
TCGCTCGATACGTTTGATGGTAC-3
0
(RC4 is located
265 bp upstream of AtHsp90-6L transcription start site),
RC2 5
0
-ACAAGCCAATAAGGTTTTAATCAGG-3
0
(RC2
starts 4 bp upstream of AtHsp90-6L transcription start site),
and RC6 5
0
-CTACCGAAATCAAAATCCACCATTC-3
(RC6 is located 169 bp downstream of AtHsp90-6L tran-
scription start site); for AtHsp90-6S, RC5 5
0
-TTCAT
TTCAATTTCCTTCATC-3
0
(RC5 starts 14 bp upstream of
AtHsp90-6S transcription start site), RC3 5
0
-TGATTGGA
TTACTTACAGTGTCACTA-3
0
(RC3 is located 78 bp
downstream of AtHsp90-6S transcription start site), rRC3 5
0
-
TAGTGACACTGTAAGTAATCCAATCA-3
0
(the same
position as for RC3), and rRC5 -5
0
-GATGAAGGAAA
TTGAAATGAA-3
0
(the same position as for RC5). Two
primers were designed in common for AtHsp90-6L and
324 Plant Mol Biol (2008) 67:323–334
123

AtHsp90-6S cDNAs: RC15, 5
0
-GATTCCATTCTCCTTG
TCTGCATAG-3
0
; RC1, 5
0
-CCGATGGTGAAACAGGTG
GCACATA-3
0
(RC15 and RC1 are located 698/483 bp and
1434/1219 bp downstream of AtHsp90-6L/AtHsp90-6S
transcription start sites, respectively). The following primers
were used for the semiquantitative reverse transcriptase-
mediated (RT) PCR experiments: AtHsp90-1,5
0
-CGCAT
GTTCAGATGGCTGATGC-3
0
and 5
0
-AGCAGAGTAGA
AACCAACACC-3
0
; for AtHsp90-5,5
0
-CTAATGGCTCC
TGCTTTGAGTAGAAG-3
0
and 5
0
- ACCAAACTGTCCG
ATCAAACCGT-3
0
; for AtHsp90-6L, RC15 5
0
-GATTCCA
TTCTCCTTGTCTGCATAG-3
0
and RC6 5
0
-CTACCGAA
ATCAAAATCCACCATTC-3
0
;forAtHsp90-6S,RC15
5
0
-GATTCCATTCTCCTTGTCTGCATAG-3
0
and RC7
5
0
-GGATCCGGTCATTTGCTCAATTGAACATG-3
0
;andfor
GFP, GFP-1 5
0
-GGAGATATAACAATGAGTAAAGGA
GAA-3
0
and GFP-2 5
0
-TTATTTGTATAGTTCATCCAT-3
0
.
For the analysis of differential expression, three rounds of RT-
PCR were conducted with two independently isolated total
RNA samples. RT-PCR was performed for 15, 20, 25, 28, 30,
and 35 cycles to determine the linearity of the PCR. The
thermal cycling parameters used for the RT-PCR for all genes
were as follows: 94°C for 15 s, 62°C for 30 s, and 72°Cfor
1 min; followed by 72°C for 10 min. As a positive control, a
540 bp AtGAPDH fragment was amplified under the same
RT-PCR conditions for a total of 20 cycles using the primer
pair 5
0
-GCAATGCATCTTGCACTACCAACTGTC-3
0
and
5
0
-CTGTGAGTAACCCCATTCATTATCSTACCA-3
0
.The
sequence identity of all RT-PCR products obtained was con-
firmed by sequencing. The amplified cDNAs were subcloned
using the pGEM-T vector system (Promega) and sequenced
using the ABI PRISM Dye Terminator Cycle Sequencing
Ready Reaction kit with fluorescent sequencing (FS) Amp-
liTaq DNA polymerase (Perkin-Elmer). Database searches
were performed on the NCBI web server by using the Basic
Local Alignment Search Tool (BLAST) network service.
Subcellular localization predictions were performed using
Predotar (http://www.inra.fr/predotar/), TargetP (http://www.
cbsdtu.dk/services/TargetP/), ChloroP (http://www.cbs.dtu.
dk/services/ChloroP/), and MitoProt (http://ihg.gsf.de/ihg/
mitoprot.html).
In vitro transcription-translation assays
In order to produce plasmids for in vitro transcription and
translation reactions, an AtHsp90-6
AUG1
open reading
frame (ORF) (584 amino acid encoding sequence) and an
AtHsp90-6
AUG2
ORF (518 amino acids encoding sequence)
were subcloned into pGEM vector (Promega). The
AtHsp90-6
AUG1
ORF and AtHsp90-6
AUG2
ORFs were
amplified with the common reverse primer TRANS-C
5
0
-TACTACTTCAAGTCCTTTCTCCAGC-3
0
and the
forward primer TRANS-1 5
0
-TCTCTTCCGAGATTTT
AGAAGTTTGC-3
0
and TRANS-2 5
0
-CAATCTAAGAA
TAGTGGGTCAT-3
0
, respectively. The plasmids were
linearized downstream of the T7 transcription terminator
and the corresponding proteins were synthesized by using
the TNT coupled transcription-translation reticulocyte
lysate system (Promega) according to the manufacturer’s
instructions.
35
S methionine-labeled proteins were ana-
lyzed by sodium dodecyl sulfate (SDS) polyacrylamide gel
electrophoresis on 12.5% gels.
GFP reporter gene constructs
The plasmid constructs for Arabidopsis transformation were
assembled as illustrated in Figs. 5 and 6. For targeting
analysis of the protein encoded by AtHsp90-5 transcripts, an
N-terminal fragment of the AtHsp90-5 gene (576 bp) was
amplified by using the following primers: 121F 5
0
-CTAAT
GGCTCCTGCTTTGAGTAGAAG-3
0
and 121R 5
0
-GTTG
TCAGCACCAAGGTCCTTGTT-3
0
. For the construct
based on the long transcript of the AtHsp90-6 gene, a 510 bp
fragment was amplified with primers RC21 5
0
-TAGAA
GTTTGCGACGATGAT-3
0
and RC15, whereas for the
corresponding construct of the short transcript, a 317 bp
fragment of AtHsp90-6 was amplified with primers RC20 5
0
-
GGTCATTTGCTCAATTGAACATG-3
0
and RC15. The
amplified fragments were subsequently fused in-frame
upstream of the smGFP sequence and cloned into the pBI101
binary vector under the control of the AtHsp90-1 promoter
(Haralampidis et al. 2002). The constructs obtained were
named as follows: pK90-5GFP, N-terminal 190 amino acid
residues of AtHsp90-5 fused to GFP; pK90-6LGFP, N-ter-
minal 164 amino acid residues of AtHsp90-6L fused to GFP;
and pK90-6SGFP, N-terminal 98 amino acid residues of
AtHsp90-6S fused to GFP. All constructs were sequenced to
check the accuracy of amplification and translational fusions
and were used to transform the Agrobacterium tumefaciens
strain GV3101::pGV2260 by the direct transfer method (An
et al. 1988). Transgenic Arabidopsis plants were obtained by
the floral dip method (Clough and Bent 1998).
Fluorescence microscopy
Localization of GFP fusions was analyzed by epifluores-
cence microscopy using an Olympus BX50 fluorescence
microscope. Images were captured with an Olympus DP71
microscope digital camera. The Arabidopsis transgenic
plants were heat shocked for 2 h at 37°C before GFP
fluorescence was analyzed. Mitochondria were visualized
with the fluorescent probe MitoTracker Orange CMTMRos
(M7510, Molecular Probes, USA) as a counterstain. The
plants were suspended in MS medium containing 25 nM
MitoTracker and incubated for 30 min at 22°C before
Plant Mol Biol (2008) 67:323–334 325
123

analyzing the fluorescence. The following filter sets were
used: for GFP, exciter HQ470/40, dichroic Q495LP, and
emitter HQ525/50; for MitoTracker Orange CMTMRos,
exciter BP 546/12, beam-splitter FT 580, emitter LP 590;
and for chlorophyll autofluorescence, 633-nm excitation
and 680 nm emission.
Results
AtHsp90-6 gene generates multiple transcripts
The full-length Hsp90-6 protein of the Arabidopsis thali-
ana is reported in the NCBI database to contain 803 amino
acid residues. However, the open reading frame of the
corresponding cDNA (NM_111652) contains four in-frame
translation initiation codons at the N-terminal region (107
aa) (AUG1-AUG4; Fig. 1). Therefore the cDNA could
potentially encode for proteins starting at any initiation
codon. The consensus sequence for translational initiation
is defined as (-6)GCCA/GCC
AUGG/A(+4) with positions
-3 and +4 being the most critical (Kozak 1986). Inspection
of the genomic sequence of AtHsp90-6 revealed that the
first two AUGs (AUG-1 and AUG-2) are located within a
sequence context matching the demand of Kozak’s rule. In
contrast, AUG-3 and AUG-4 are located within a poor
Kozak configuration to be considered as initiation codons.
These observations led us to investigate whether AtHsp90-6
could be a multifunctional gene, coding for various forms of
Hsp90 by alternative use of transcription/translation initia-
tion codons.
Primer extension analysis was performed to determine
the potential transcription initiation sites of AtHsp90-6. The
results indicated that two major transcript ends can be
detected in A. thaliana: one mapping 228 nucleotides
upstream of the first translation start site (AUG-1) and one
mapping 210 nucleotides upstream of the second
translation start site (AUG-2) (Fig. 2). Consequently, it is
very likely that the AtHsp90-6 gene has multiple tran-
scriptional start sites. The first exon of the longer transcript
(designated AtHsp90-6L) is identical to that deposited in
the Arabidopsis genome database. The ATG start codon, of
the shorter transcript (designated AtHsp90-6S), is located
within the second exon of the AtHsp90-6L. The 5
0
-region of
the AtHsp90-6L sequence contains a predicted intron of
137 bp (Fig. 2a). The DNA sequence conservation rule
near the exon–intron boundaries is fulfilled, with GT
located at the initiation and AG at the termination sites of
the intron, respectively. To corroborate the presence of two
different mRNA populations corresponding to AtHsp90-6L
and AtHsp90-6S, RT-PCR analysis was performed using
forward and reverse oligonucleotide primers specific for
the 5
0
-UTR (RC4, RC2 and RC6), intron 1 (RC5, RC3 and
rRC3), exon 4 (RC15), and exon 12 (RC1), as shown in
Fig. 3. DNA fragments of different sizes were amplified
and the two transcript populations were confirmed by
nucleotide sequencing of the RT-PCR products. The cor-
responding cDNAs did not contain any intron sequences.
As anticipated, no amplification products were obtained
using primer combinations RC4-RC1 and RC5-RC1 while
the primer combination RC3-RC1 produced the predicted
fragment. Since both RC3 and RC5 primers are located
within the first intron and the latter covers the determined
initiation of transcription (Fig. 2), the existence of the
smaller mRNA is expected. To rule out the possibility that
the PCR products originated from contaminating genomic
DNA, additional RT-PCR reactions were performed.
Reverse primer rRC3 (located in the first intron) together
with the sense primers RC6 or RC2 did not produce any
amplification fragments. Sequencing of both cDNAs
deriving from the long and short transcripts showed that
they were identical except for the 5
0
ends. We therefore
concluded that AtHsp90-6 produces at least two different
mRNAs encoding AtHsp90-6L (799 amino acids from 20
exons) and AtHsp90-6S (733 amino acids from 19 exons)
proteins. It should be noted that the AtHsp90-6 gene in the
NCBI database possesses 21 exons. However a careful
comparison of the genomic sequence to the cDNA
sequences obtained in the course of the present study
revealed that AtHsp90-6 consists of 20 exons. This error is
due to a misannotation of the intron/exon boundaries for
exons 16 and 17.
In silico analysis of the 1 kb region upstream of the
AtHsp90-6 gene transcription start sites revealed the pres-
ence of a heat-shock element (HSE) consisting of three
perfect and one imperfect core units (cTTCaaaTCca-
GAAgcTTCg) (Fig. 2A). Two CCAAT sequences were
identified near the putative HSE. It has been reported that
CCAAT-box sequences (representing the binding sites for
the C/EBP transcription factors) act cooperatively with
Fig. 1 Nucleotide context of AUGs contained within the first 107
amino acids of the A. thaliana Hsp90-6. Upper: positions of
alternative translational initiation codons (AUG1-AUG4) are desig-
nated by arrowheads. Lower: table showing the nucleotide sequence
flanking the four AUGs (indicated in bold) and the length of the
resulting ORFs
326 Plant Mol Biol (2008) 67:323–334
123

HSEs (Rieping and Schoffl 1992; Prasinos et al. 2004).
Interestingly, no perfect TATA boxes were identified.
Further inspection of the promoter region revealed the
presence of potential binding sites for various transcription
factors including activating-enhancer protein 1 (AP-1),
mitochondrial stress response element (MSR-like), and
stress response element (STRE) (data not shown). These
cis-elements play important roles in regulating expression
of genes in response to heat shock, heavy metal, dehy-
dration, low temperature, light, and pathogen elicitors
(Haralampidis et al. 2002; Zhao et al. 2002; Takahashi
et al. 2003). However, the biological significance of the
-233 tacaagccaataaggttttaatcaggattatccttaattaatgtattcatacttaccgttca
caaaataattatgagaaaaccaaataccggaattacccttcaaatccagaagcttcggcttt
ggcctaaaaaccctactaaaccccaat
cgtctctcttcttagctcagagtgctaccgaaatc
aaaatccaccattcttttctcttccgagattttagaagtttgcgacgATGATCAGGCTCTCT
AAGCGCTCCGTCTCTACCCTCCTACGCTCCGGTAACCAAAGCTTCCGTATCTCGCCGCCGCA
GCTTCCACCTCCCGTTCTTCCCCATCTGCCACGgtattcatttcaatttccttcatctatcc
ctcgctttgggttttggtttctcggttcagtggccaattggctttcgtatgataggcgaaga
gaattgattggattacttacagtgtcactattttccccccttttagGATGTCAAGAGAAGTG
ACACTGAATCGAGATGGTACTCATCTTTAACCAATGGACAATCTAAGAATAGTGGGTCATTT
GCTCAATTGAACATGAAAACCAATTGGTTTATGGGATACCGGAACGAATCCAGTGCAGCAGC
ATCAGATTCTTCCTCACAAGCTCCTCCACCGGCTGAGAAATTCGAGTATCAAGCTGAAgtac
+488 gtacgatctgtgtgtttagatgatttcttgttttgtgat
a
58 kD
65 kD
cb
AtHsp90-6S
AtHsp90-6L
AtHsp90-6L
T C A G
T
A
C
A
A
G
C
C
A
A
T
A
A
231bp
AtHsp90-6S
T C A G
C
A
A
T
T
T
C
C
T
T
C
A
T
233bp
Fig. 2 Alternative transcription start sites of the Arabidopsis
AtHsp90-6 gene. (a) Partial sequence of the 5
0
region AtHsp90-6
gene showing the two in-frame ATG codons (indicated in boldface).
The first intron of the Hsp90-6L transcript is represented in lowercase
italics, and boundaries in bold letters. Potential start sites for
AtHsp90-6L and AtHsp90-6S transcripts are highlighted black.
Potential heat shock element (HSE) is indicated by black dots.
CCAAT boxes are underlined. (b) The transcription initiation sites of
the AtHsp90-6L and AtHsp90-6S mRNAs were determined by primer
extension analysis. The same primers (PRET-1 and -2) were used for
sequencing of an appropriate fragment. The products of extension and
sequencing analysis were electrophoresed on the same gel. In the
vicinity of the alternative transcription start sites, the nucleotide
sequence is shown to the left of each gel. Arrows show the nucleotide
positions of the 5
0
ends of the transcripts relative to PRET-1 and
PRET-2. (c) The AtHsp90-6L and AtHsp90-6S transcripts were
translated in a rabbit reticulocyte lysate system using [
35
S]Met.
Samples were separated on a 12% SDS polyacrylamide gel. Bands
corresponding to AtHsp90-6L and AtHsp90-6S are indicated by
arrows. The estimated molecular masses are also indicated.
Plant Mol Biol (2008) 67:323–334 327
123

putative elements within the promoter region of the
AtHsp90-6 remains to be verified experimentally.
Since our results suggested that the annotated Met codon
(AUG-1) might not serve as the only site for the initiation
of translation, we investigated whether this Met codon was
dispensable for translation of the gene. Given that every
ATG codon is not necessarily used for translation initia-
tion, we attempted to translate in vitro the two AtHsp90-6
coding sequences. Open reading frames encoding 584 or
518 amino acids of the AtHsp90-6L or AtHsp90-6S cDNA,
respectively, were cloned into the pGEM vector and the
resulting polypeptide products were
35
S methionine labeled
in a TNT Coupled Reticulocyte Lysate System. As shown
in Fig. 2c, two translation products of 65 or 58 kDa were
produced from the long or the short AtHsp90-6 template,
respectively. These results suggested that both ATGs are
suited for translation initiation, as predicted by the
presence of typical Kozak sequences. We therefore propose
that the Hsp90-6 isoforms were not generated by alterna-
tive splicing but resulted from the utilization of alternative
transcription initiation sites.
Subcellular localization of AtHsp90-6L:GFP
and AtHsp90-6S:GFP fusion proteins
The majority of nuclear-encoded organellar proteins are
translated by cytosolic ribosomes and directed to the
appropriate organelle by the N-terminal signal peptides.
Sequence comparison between members of the AtHsp90
protein family revealed that AtHsp90-6L, AtHsp90-6S, and
AtHsp90-5 contain a highly variable N-terminal region
composed of 28–94 amino acids (Fig. 4). In the case of
AtHsp90-6L, the targeting peptide is predicted to be 31
amino acids long as determined by in silico analysis
Fig. 3 Schematic representation of the strategy used to confirm the
existence of two different AtHsp90-6 mRNAs. The relative position
and the orientation of primers used to amplify the AtHsp90-6L and
AtHsp90-6S transcripts are shown by arrows. A common reverse
primer (RC15) was used in combination to different sense primers in
order to obtain a consistent 3
0
end and varying ends at the 5
0
. The
amplification results were also validated by another common reverse
primer (RC1). The primers RC3, RC5, and rRC3 were designed inside
the first intron of AtHsp90-6L, hence fragments specific to AtHsp90-
6S mRNA were amplified. For further details see the ‘Materials and
methods’ section. The schematic representation has not been drawn to
scale
Fig. 4 Sequence alignment of the N-terminal regions of AtHsp90-5,
AtHsp90-6L, and AtHsp90-6S. Identical or highly similar residues are
highlighted black or grey, respectively. Dashes indicate gaps.
Numbers indicate the amino acid position. Residues in white boxes
show the predicted transient peptide of the organellar Hsps.
Sequences used to generate the translational GFP fusion constructs
are underlined. Accession numbers: AtHsp90-5, NP 849932;
AtHsp90-6, NP187434
328 Plant Mol Biol (2008) 67:323–334
123

(MitoProt). The intracellular prediction program TargetP
indicated that AtHsp90-6L could be localized to both
mitochondria and chloroplasts, whereas Predotar predicted
only mitochondrial localization. Interestingly, in the case
of AtHsp90-6S, in silico protein targeting analysis indi-
cated that the translation initiation at the AUG-2 start
codon did not contain any organelle target sequence. To
determine the intracellular targeting of AtHsp90-6L and
AtHsp90-6S proteins in planta, pK90-6LGFP and pK90-
6SGFP reporter constructs were used for Arabidopsis
transformation (Fig. 5a). Arabidopsis transgenic lines
expressing the fusion proteins were established and ana-
lyzed by epifluorescence microscopy. Both pK90-6LGFP
and pK90-6SGFP fusion proteins were detected in
numerous spherical bodies of root-hair cells (Fig. 5b).
Staining with the mitochondria-specific dye MitoTracker
Orange (Molecular Probes) revealed co-localization with
the GFP fluorescent signals, confirming that pK90-6LGFP
and pK90-6SGFP fusion proteins were delivered into
mitochondria. In control plants, only red and no green
fluorescence was detected (data not shown). To investigate
whether any of the two fusion proteins had a dual targeting
to both mitochondria and chloroplasts, isolated Arabidopsis
leaf protoplasts were analyzed. Using the autofluorescence
of chlorophyll as a marker, it was clear that the GFP signal
did not co-localize with the red chlorophyll autofluores-
cence (Fig. 5b lower panel). The above results indicate that
the N-terminal sequences present in pK90-6LGFP and
pK90-6SGFP include a mitochondrial targeting sequence
that is necessary and sufficient to transport the fusion
proteins into mitochondria.
The predicted transit peptide targets AtHsp90-5
exclusively to chloroplasts
There are many and varied ways by which cross-com-
partment targeting is achieved by nuclear-encoded
organellar proteins including N-terminal and internal signal
Fig. 5 In vivo targeting of
AtHsp90-6L::GFP and
AtHsp90-6S::GFP fusion
proteins in Arabidopsis.(a)
Depiction of the AtHsp90-6
fusion constructs used for stable
transformation of Arabidopsis
thaliana plants. Expression was
controlled by the AtHsp90-1
promoter. (b) Root-hair cells
(upper and middle panels) and
isolated protoplasts (lower
panel) of stable A. thaliana
transformants expressing
AtHsp90-6L::GFP or AtHsp90-
6S::GFP fusion proteins.
Mitotracker Orange (red
channel) was used as a
mitochondrial fluorescent
marker in root-hair cells. The
red channel in protoplasts shows
the autofluorescence of
chloroplasts. Overlay panels are
merged images of the GFP and
Mitotracker or GFP and
chloroplast autofluorescence
Plant Mol Biol (2008) 67:323–334 329
123

sequences. Computer-assisted analysis for localization
signals predicted that the AtHsp90-5 protein has a
60-amino-acid transit peptide (TP) (Fig. 4). The intracel-
lular prediction program TargetP indicated that the
AtHsp90-5 could be localized to both chloroplasts and
mitochondria, while Predotar predicted only chloroplastic
localization. To independently verify the plastidic locali-
zation of AtHsp90-5 in planta, a pK90-5GFP fusion
construct was used to generate stably transgenic Arabid-
opsis plants. Targeting of the encoded fusion protein was
monitored using epifluorescence microscopy presented in
Fig. 6. The green fluorescence of GFP clearly co-localized
with the red autofluorescence of chlorophyll (Fig. 6b,
middle panels), demonstrating that the protein was effi-
ciently targeted to chloroplasts. To investigate whether the
fusion protein was also targeted to mitochondria, Arabid-
opsis root cells were analyzed. Using Mitotracker as a
fluorescent marker for mitochondria, it was clear that the
detected GFP signal in plastids of root cells did not overlap
with the red Mitotracker fluorescence signal (Fig. 6b lower
panel). In transgenic lines harboring the GFP control con-
struct (pK90-1GFP), the fluorescence was observed in the
cytoplasm, corroborating the principally cytosolic locali-
zation reported for wild-type GFP (Fig. 6b, upper panel).
Exogenously Hsp90 mRNAs resemble
the corresponding endogenous mRNAs
In order to investigate how the abundance of AtHsp90-5,
AtHsp90-6L, and AtHsp90-6S was regulated under control
or heat-stress conditions, we determined the levels of the
corresponding transcripts using semiquantitative RT-PCR
on RNA isolated from 28-day-old plants. AtHsp90-1
mRNA levels were also monitored (Fig. 7a). Whereas a
strong GAPDH-derived band was observed, neither
AtHsp90-5- nor AtHsp90-6-derived transcript was detect-
able after 25 cycles of PCR (data not shown). After 30
cycles of PCR amplification, AtHsp90-6S mRNA was
detected under heat-stress conditions whereas after 35
cycles, both AtHsp90-6L and AtHsp90-6S mRNAs were
amplified under control or heat-shock conditions. The
results indicate that the expression profile of both transcript
forms is similar; however AtHsp90-6S was expressed at
higher levels under the conditions tested. AtHsp90-5
Fig. 6 In vivo targeting of the
AtHsp90-5::GFP fusion protein
into Arabidopsis chloroplasts.
(a) Depiction of pK90-1GFP
(control) and pK90-5GFP fusion
constructs used for stable
transformation of Arabidopsis
thaliana plants. Expression was
controlled by the AtHsp90-1
promoter. (b) Plant tissues were
inspected by differential
interference contrast (DIC)
microscopy and fluorescence
signal was detected by
epifluorescent microscopy. The
red channel shows
autofluorescence of chlorophyll
(upper three panels) in
photosynthetic tissues and
Mitotracker fluorescence (lower
panel) in roots. The merged
images of GFP and chlorophyll
autofluorescence confirm the
plastidic targeting of the
AtHsp90-5. In roots, GFP
plastidial localization did not
co-localize with Mitotracker
stained mitochondria. GFP
alone (pK90-1GFP) shows the
characteristic fluorescence in
the cytosol
330 Plant Mol Biol (2008) 67:323–334
123

expression levels were low under control or heat-shock
conditions. As shown previously, AtHsp90-5 mRNA levels
were almost undetectable in mature plants, suggesting that
the expression of AtHsp90-5 is developmentally regulated
(Cao et al. 2003).
To determine whether the expression pattern of the
fused genes in the transgenes harboring the constructs
pK90-1:GFP, pK90-5:GFP, pK90-6L:GFP or pK90-
6S:GFP resembled that of the corresponding endogenous
AtHsp90 mRNA levels, RT-PCR analysis was performed.
Twenty-eight-day-old transgenic Arabidopsis plants har-
boring a single transgene copy were exposed to 37°C for 0
or 2 h. RNA was reverse-transcribed from a pool of ten
independent transgenic lines and the expression profiles of
the AtHsp90-1, AtHsp90-5, AtHsp90-6L, and AtHsp90-6S
were determined. Transcription patterns of the GFP gene
constructs and the corresponding endogenous genes were
similar for both GFP transgenes and wild-type plants
(Fig. 7a, b). However, in transgenic plants, the GFP fusion
transcript levels were higher than the endogenous under
heat-shock conditions, suggesting a stronger interaction of
the AtHsp90-1 promoter to Hsfs and/or a difference in
mRNA stability (Salvador et al. 2004).
Discussion
In Arabidopsis the organellar Hsp90 subfamily of proteins
consists of two members, designated AtHsp90-5
(At2g04030) and AtHsp90-6 (At3g07770). In this paper we
show that AtHsp90-5 encodes a protein localized specifi-
cally to the chloroplasts but not to the mitochondria, while
AtHsp90-6 encodes two different gene products, both
localized into mitochondria. Chloroplasts and mitochondria
are remnants of free-living prokaryotes that lost their
autonomy during evolution by establishing an endosym-
biotic relationship with their host cells. However, it has
been suggested that none of the organellar Hsp90s were
derived from endosymbiotic events during eukaryote evo-
lution, in marked contrast to Hsp60 and Hsp70 (Stechmann
and Cavalier-Smith 2004).
Alternative transcript initiation and alternative splicing
are two notable mechanisms for subcellular localization
(Silva-Filho 2003). A number of alternative transcription
events have been reported in plants (Obara et al. 2002;
Wachter et al. 2005; Parsley and Hibberd 2006). The
results obtained by primer extension analysis are consistent
with the notion that the AtHsp90-6 gene generates at least
two transcripts, allowing the production of a second, pre-
viously undetected mRNA that encodes a novel Hsp90
isoform (named AtHsp90-6S). The first exon of the long
transcript is identical to that predicted for the AtHsp90-6
gene deposited in the genome database. The alternative
transcriptional initiation start site of the short transcript was
identified within the first intron of the long form. The
occurrence of the AtHsp90-6L and AtHsp90-6S mRNAs
was demonstrated by cloning of the corresponding cDNAs.
To evaluate whether the observed transcript heterogeneity
is a unique feature of A. thaliana, computational search for
expressed sequence tags (ESTs) from different plant
species was performed. The presence of two Hsp90-6-like
transcript populations derived from the mtHsp90
(Os12g32986) gene was identified in the rice genome
sequence database (http://www.gramene.org/Oryza_sativa/
geneview). The overall genomic structure of the
Os12g32986 rice gene resembles that of the AtHsp90-6 gene
from Arabidopsis. Alignment of the amino acid sequences of
AtHsp90-6 and the Os12g32986 protein product revealed
84.4% homology and 91.5% similarity at the amino acid
level. Some collinearity exists between the rice and
AtHsp90-1
AtHsp90-6L
22
°
C 37
°
C 22
°
C 37
°
C 22
°
C 37
°
C 22
°
C 37
°
C
AtHsp90-6S AtHsp90-5 AtHsp90-1
GAPDH
X30
X35
GFP
22
°
C 37
°
C
endogenous
22
°
C 37
°
C
AtHsp90-5
AtHsp90-6L
AtHsp90-6S
GAPDH
a
b
Fig. 7 Expression patterns of the organellar AtHsp90 genes (a)
Semiquantitative RT-PCR analysis of endogenous AtHsp90-1,
AtHsp90-5, AtHsp90-6L, and AtHsp90-6S expression levels under
normal or heat-stress conditions as detailed in the ‘Materials and
methods’ section. PCR amplification was carried out for 30 or 35
cycles. (b) Semiquantitative RT-PCR analysis of Hsp90 genes
(endogenous) or GFP (GFP) driven by the AtHsp90-1 promoter
alone or in combination with the respective transit peptide sequence.
RNA was isolated from 4-week-old Arabidopsis transgenes harboring
pK90-1GFP (control), pK90-5GFP, pK90-6LGFP, and pK90-6SGFP
under normal or heat-stress conditions. In both (a) and (b) the
Arabidopsis GAPDH gene was used as an internal control for
normalization
Plant Mol Biol (2008) 67:323–334 331
123

Arabidopsis genomes and initial comparative microarray
analysis has revealed some basic similarities between the
two transcriptomes (Ma et al. 2005). The above observation
indicates that an alternative transcription/translational pat-
tern may be conserved in organellar Hsp90 genes of
monocotyledonous and dicotyledonous plants. However, the
biological significance of this phenomenon in plants has yet
to be determined. This different transcriptional regulation
may facilitate spatial and temporal regulation of Hsp90 gene
expression, fine-tune Hsp90-6 protein function at different
stages of development, and/or provide adaptive responses to
the environment. In animal systems, alternative transcription
of the Hsp70-1 gene might be primarily linked to the
pathophysiology of human depression or secondary to spe-
cific pleiotropic effects (Shimizu et al. 1999). It has been
postulated that multiple transcription initiation sites and
alternative splicing events are frequently used to create
diversity and flexibility in the regulation of gene expression.
Global analysis of alternative splicing events has revealed
that about 12–20% of Arabidopsis nuclear genes produce
multiple RNAs (Wang and Brendel 2006). In the case of the
SHD gene (At4g24190), encoding an ER-targeted Hsp90,
genome-wide analysis of pre-mRNA splicing in Arabidopsis
revealed the presence of different transcript populations due
to alternative terminal (AT) exon splicing (Iida et al. 2004).
Maize mitochondrial Hsp22 has also been shown to be
subject to alternative intron splicing (Lund et al. 2001).
In silico studies on promoter cis-elements could allow
the functional dissection of the Arabidopsis mitochondrial
Hsp90 gene. In mammals when cells are exposed to stress
conditions, CHOP (GADD153), which belongs to a bZIP
transcription factor family, has been shown to activate
nuclear genes encoding mitochondrial stress proteins (Zhao
et al. 2002). The modification of nuclear gene expression
in response to changes in mitochondrial status has been
termed mitochondrial retrograde regulation (MRR).
Although little is known about the mechanism of com-
munication between mitochondria and the nucleus in
plants, it has been reported that MRR can occur during heat
stress (Yu et al. 2001; Rhoads et al. 2005). An MSR-like
cis-acting regulatory element was found in the promoter
region of the Arabidopsis AtHsp90-6 gene, suggesting that
Hsp90-6 could potentially be regulated by an Arabidopsis
CHOP-like transcription factor.
Our investigation has indicated differential accumula-
tion of AtHsp90-6L and AtHsp90-6S transcripts in heat-
stressed Arabidopsis plants, implying that AtHsp90-6 may
fulfill multiple roles. The expression of AtHsp90-6 or
AtHsp90-5 is strongly induced in response to a number of
stress treatments, including heat, arsenite, and light treat-
ment (Milioni and Hatzopoulos 1997; Cao et al. 2003). A
range of defense strategies such as protection of existing
mitochondria’s matrix enzymes by the synthesis of soluble
protein-folding molecular chaperones such as Hsp90
appears to exist in plants, which could potentially help the
organism to minimize the damage generated by various
stress conditions (Taylor et al.
2005). However, it is
plausible that developmental processes could be involved
in the regulation of AtHsp90-5 (Cao et al. 2003)or
AtHsp90-6 expression.
The subcellular localization analysis provided evidence
for an organellar role of AtHsp90-5, AtHsp90-6L, and
AtHsp90-6S proteins. Previous work has shown that the
in vitro translation product of AtHsp90-5 was imported into
the pea chloroplast stroma compartment (Cao et al. 2003).
Herein, we demonstrated in planta the chloroplastic
localization of the protein by using a partial AtHsp90-5
cDNA fused to GFP. Furthermore, we showed that
AtHsp90-6L and AtHsp90-6S were targeted into the same
cellular compartment. Both proteins could import the fused
GFP into mitochondria. Although AtHsp90-6L signal
peptide has the characteristics of a mitochondrial pre-
sequence, AtHsp90-6S did not contain any mitochondrial
localization signal. This indicates that the presence of
AtHsp90-6S in mitochondria may be due to the existence
of an unconventional mitochondrial-sorting signal. Inde-
pendent support of Hsp90-6 localization into Arabidopsis
mitochondria is provided by proteomic studies using ana-
lytical methods (Heazlewood et al. 2004; Millar et al.
2005). However, an in-depth inspection of the pK90-
6SGFP micrographs showed that a number of green fluo-
rescent spherical bodies did not stain with the MitoTracker
Orange. Since the Hsp90 function is essential for basic
cellular activities, it is anticipated that this chaperone could
be present in most compartments (organelles) of the cell. In
plants, mitochondria are known to contain the chaperones
Hsp70, Cnp60, small Hsps, and a form of Hsp100 (Clp),
which are involved in protein import, remodeling, folding,
and assembly (Sigler et al. 1998; Merlin et al. 1999; Sun
et al. 2002; Peltier et al. 2004). However, mitochondrial
Hsp90 has been thus far characterized only in animals but
not in yeast or plants. In mammals, the Hsp90-related
protein tumor necrosis factor-associated protein 1 (TRAP1)
is primarily a mitochondrial matrix protein. However, it
has also been localized to various extramitochondrial sites
such as secretory granules, nuclei, and at the cell surface
(Cechetto et al. 2000). Furthermore, in differentiating
Dictyostelium prespore cells TRAP1 is located in the
prespore-specific vacuole (PSV), a unique cell-type-spe-
cific organelle (Yamaguchi et al. 2005). TRAP1 has been
implicated in protecting mitochondria against damaging
stimuli via a decrease of reactive oxygen species (ROS)
production (Im et al. 2007). Hsp90 chaperone does not act
to fold nonnative proteins but rather binds to substrate
proteins at a late stage of folding (Pearl and Prodromou
2002) and accompanies proteins involved in signal
332 Plant Mol Biol (2008) 67:323–334
123

transduction (Zhao et al. 2005). Glucocorticoid receptors
(GRs) have been shown to be located within the mito-
chondria, eliciting apoptosis in some animal cell types
(Sionov et al. 2006). It is plausible that GR is transported
to the mitochondria by a heat-shock protein, as its ligand
binding (558–580) domain overlaps with one of its Hsp90
binding sites (Schaaf and Cidlowski 2002). Although it is
alluring to speculate that the A. thaliana mtHsp90 may
function in a mitochondrial stress-response mechanism that
acts to prevent cellular damage and to re-establish cellular
homeostasis or sustain the function of proteins participat-
ing in intracellular signal transduction networks, further
studies are required to relate this molecular chaperone to
specific targets.
Acknowledgements We would like to thank Elli Hatzistavrou for
technical assistance. This work was partly supported by a grant to PH
from the GSRT, Greece (PENED 01/148) and Pythagoras I.
References
An G, Edbert PR, Mitra A, Ha SB (1988) Binary vectors. In: Gelvin
SB, Shilperoort RA, Verma DPS (eds) Plant molecular biology
manual. Kluwer Academic, Dordrecht, The Netherlands
Boter M, Amigues B, Peart J, Breuer C, Kadota Y, Casais C, Moore
G, Kleanthous C, Ochsenbein F, Shirasu K,Guerois R (2007)
Structural and functional analysis of SGT1 reveals that its
interaction with HSP90 is required for the accumulation of Rx,
an R protein involved in plant immunity. Plant Cell 19:
3791–3804
Cao D, Froehlich JE, Zhang H, Cheng C-L (2003) The chlorate-
resistant and photomorphogenesis-defective mutant cr88
encodes a chloroplast-targeted HSP90. Plant J 33:107–118
Cechetto JD, Gupta RS (2000) Immunoelectron microscopy provides
evidence that tumor necrosis factor receptor-associated protein 1
(TRAP-1) is a mitochondrial protein which also localizes at
specific extramitochondrial sites. Exp Cell Res 260:30–39
Chen B, Zhong D, Monteiro A (2006) Comparative genomics and
evolution of the HSP90 family of genes across all kingdoms of
organisms. BMC Genomics 7:156
Clough SJ, Bent AF (1998) Floral dip: a simplified method for
Agrobacterium-mediated transformation of Arabidopsis thali-
ana. Plant J 16:735–743
Doelling JH, Pikaard CS (1993) Transient expression in Arabidopsis
thaliana protoplasts derived from rapidly established cell
suspension cultures. Plant Cell Rep 12:241–244
Emelyanov V (2002) Phylogenetic relationships of organellar Hsp90
homologs reveal fundamental differences to organellar Hsp70
and Hsp60 evolution. Gene 299:125–133
Felts SJ, Owen BAL, Nguyen P, Trepel J, Donner DB, Toft DO
(2000) The hsp90-related protein TRAP1 is a mitochondrial
protein with distinct functional properties. J Biol Chem
275:3305–3312
Haralampidis K, Milioni D, Rigas S, Hatzopoulos P (2002) Combi-
natorial Interaction of cis elements specifies the expression of the
Arabidopsis AtHsp90-1 gene. Plant Physiol 129:1138–1149
Heazlewood JL, Tonti-Filippini JS, Gout AM, Day DA, Whelan J,
Millar AH (2004) Experimental analysis of the Arabidopsis
mitochondrial proteome highlights signaling and regulatory
components, provides assessment of targeting prediction
programs, and indicates plant-specific mitochondrial proteins.
Plant Cell 16:241–256
Iida K, Seki M, Sakurai T, Satou M, Akiyama K, Toyoda T,
Konagaya A, Shinozaki K (2004) Genome-wide analysis of
alternative pre-mRNA splicing in Arabidopsis thaliana based on
full length cDNA sequences. Nucl Acids Res 32:5096–5103
Im CN, Lee JS, Zheng Y, Seo JS (2007) Iron chelation study in a
normal human hepatocyte cell line suggests that tumor necrosis
factor receptor-associated protein 1 (TRAP1) regulates produc-
tion of reactive oxygen species. J Cell Biochem 100:474–486
Ishiguro S, Watanabe Y, Ito N, Nonaka H, Takeda N, Sakai T,
Kanaya H, Okada K (2002) SHEPHERD is the Arabidopsis
GRP94 responsible for the formation of functional CLAVATA
proteins. EMBO J 21:898–908
Joab I, Radanyi C, Renoir M, Buchou T, Catelli MG, Binart N,
Mester J, Baulieu EE (1984) Common non hormone binding
component in non transformed chick oviduct receptors of four
steroid hormones. Nature 308:850–853
Kozak M (1986) Point mutations define a sequence flanking the AUG
initiator codon that modulates translation by eukaryotic ribo-
somes. Cell 44:283–292
Krishna P, Gloor G (2001) The Hsp90 family of proteins in
Arabidopsis thaliana. Cell Stress Chaperones 6:238–246
Lund AA, Rhoads DM, Lund AL, Cerny RL, Elthon TE (2001)
In vivo modifications of the maize mitochondrial small heat
stress protein, HSP22. J Biol Chem 276:29924–29929
Ma L, Chen C, Liu X, Jiao Y, Su N, Li L, Wang X, Cao M, Sun N,
Zhang X, Bao J, Li J, Pedersen S, Bolund L, Zhao H, Yuan L,
Wong GK, Wang J, Deng XW, Wang J (2005) A microarray
analysis of the rice transcriptome and its comparison to
Arabidopsis. Genome Res 15:1274–1283
Masuda Y, Shima G, Aiuchi T, Horie M, Hori K, Nakajo S, Kajimoto
S, Shibayama-Imazu T, Nakaya K (2004) Involvement of tumor
necrosis factor receptor-associated protein 1 (TRAP1) in apop-
tosis induced by beta-hydroxyisovalerylshikonin. J Biol Chem
279:42503–42515
Merlin A, Voos W, Maarse AC, Meijer M, Pfanner N, Rassow J
(1999) The J-related segment of Tim44 is essential for cell
viability: A mutant Tim44 remains in the mitochondrial import
site, but inefficiently recruits mtHsp70 and impairs protein
translocation. J Cell Biol 145:961–972
Milioni D, Hatzopoulos P (1997) Genomic organization of hsp90
gene family in Arabidopsis. Plant Mol Biol 35:955–961
Millar AH, Heazlewood JL, Kristensen BK, Braun H-P, Moller IM
(2005) The plant mitochondrial proteome. Trends Plant Sci
10:36–43
Obara K, Sumi K, Fukuda H (2002) The use of multiple transcription
starts causes the dual targeting of Arabidopsis putative monode-
hydroascorbate reductase to both mitochondria and chloroplasts.
Plant Cell Physiol 43:697–705
Parsley K, Hibberd JM (2006) The Arabidopsis
PPDK gene is
transcribed from two promoters to produce differentially
expressed transcripts responsible for cytosolic and plastidic
proteins. Plant Mol Biol 62:339–349
Peltier JB, Ripoll DR, Friso G, Rudella A, Cai Y, Ytterberg J,
Giacomelli L, Pillardy J, Van Wijk KJ (2004) Clp protease
complexes from photosynthetic and non-photosynthetic plastids
and mitochondria of plants, their predicted 3-D structures and
functional implications. J Biol Chem 279:4768–4781
Pearl LH, Prodromou C (2002) Structure, function and mechanism of
the Hsp90 molecular chaperone. Adv Protein Chem 59:157–186
Prasinos C, Krampis K, Samakovli D, Hatzopoulos P (2004) Tight
regulation of expression of two Arabidopsis cytosolic Hsp90
genes during embryo development. J Exp Bot 56:633–644
Queitsch C, Sangster TA, Lindquist S (2002) Hsp90 as a capacitor of
phenotypic variation. Nature 417:618–624
Plant Mol Biol (2008) 67:323–334 333
123

Rhoads DM, White SJ, Zhou Y, Muralidharan M, Elthon TE (2005)
Altered gene expression in plants with constitutive expression of
a mitochondrial small heat shock protein suggests the involve-
ment of retrograde regulation in the heat stress response. Physiol
Plant 123:435–444
Rieping M, Schoffl F (1992) Synergistic effect of upstream
sequences, CCAAT box elements, and HSE sequences for
enhanced expression of chimeric heat-shock genes in transgenic
tobacco. Mol Gen Genet 231:226–232
Rutherford SL, Lindquist S (1998) Hsp90 as a capacitor for
morphological evolution. Nature 396:336–342
Salvador ML, Suay L, Anthonisen IL, Klein U (2004) Changes in the
5
0
-untranslated region of the rbcL gene accelerate transcript
degradation more than 50-fold in the chloroplast of Chlamydo-
monas reinhardtii. Curr Genet 45:176–182
Samakovli D, Thanou A, Vamas C, Hatzopoulos P (2007) Hsp90
canalizes developmental perturbation. J Exp Bot 58:3513–3524
Sangster TA, Queitsch C (2005) The HSP90 chaperone complex, an
emerging force in plant development and phenotypic plasticity.
Curr Opin Plant Biol 8:86–92
Schaaf JM, Cidlowski JA (2002) Molecular mechanisms of gluco-
corticoid action and resistance. J Steroid Biochem Mol Biol
83:37–48
Schatz G, Dobberstein B (1996) Common principles of protein
translocation across membranes. Science 271:1519–1526
Sepehrnia B, Paz IB, Dasgupta G, Momand J (1996) Heat shock
protein 84 forms a complex with mutant p53 protein predom-
inantly within a cytoplasmic compartment of the cell. J Biol
Chem 271:15084–15090
Shimizu S, Nomura K, Ujihara M, Demura H (1999) An additional
exon of stress-inducible heat shock protein 70 gene (HSP70–1).
Biochem Biophys Res Commun 257:193–198
Sigler PB, Xu Z, Rye HS, Burston SG, Fenton WA, Horwich AL
(1998) Links, structure and function in GroEL-mediated protein
folding. Annu Rev Biochem 67:581–608
Silva-Filho MC (2003) One ticket for multiple destinations: dual
targeting of proteins to distinct subcellular locations. Curr Opin
Plant Biol 6:589–595
Sionov RV, Cohen O, Kfir S, Zilberman Y, Yefeno E (2006) Role of
mitochondrial glucocorticoid receptor in glucocorticoid-induced
apoptosis. J Exp Med 203:189–203
Song HY, Dunbar JD, Zhang Y X, Guo D, Donner DB (1995)
Identification of a protein with homology to hsp90 that binds the
type 1 tumor necrosis factor receptor. J Biol Chem 270:
3574–3581
Stechmann A, Cavalier-Smith T (2004) Evolutionary origins of
Hsp90 chaperones and a deep paralogy in their bacterial
ancestors. J Eukaryot Microbiol 51:364–373
Sun W, Van Montagu M, Verbruggen N (2002) Small heat shock
proteins and stress tolerance in plants. Biochim Biophys Acta
1577:1–9
Takahashi A, Casais C, Ichimura K, Shirasu K (2003) HSP90
interacts with RAR1 and SGT1 and is essential for RPS2-
mediated disease resistance in Arabidopsis. Proc Natl Acad Sci
USA 100:11777–11782
Taylor NL, Heazlewood JL, Day DA, Millar AH (2005) Differential
impact of environmental stresses on the pea mitochondrial
proteome. Mol Cell Proteomics 4:1122–1133
Wachter A, Wolf S, Steininger H, Bogs J, Rausch T (2005)
Differential targeting of GSH1 and GSH2 is achieved by
multiple transcription initiation: implications for the compart-
mentation of glutathione biosynthesis in the Brassicaceae. Plant
J 41:15–30
Wang BB, Brendel V (2006) Genome wide comparative analysis of
alternative splicing in plants. Proc Natl Acad Sci USA
103:7175–7180
Wegele H, Muller L, Buchner J (2004) Hsp70 and Hsp90—a relay
team for protein folding. Rev Physiol Biochem Pharmacol
151:1–44
Wilhelmsson A, Cuthill S, Denis M, Wikstrom AC, Gustafsson JA,
Poellinger L (1990) The specific DNA binding activity of the
dioxin receptor is modulated by the 90 kd heat shock protein.
EMBO J 9:69–76
Yamaguchi H, Morita T, Amagai A, Maeda Y (2005) Changes in
spatial and temporal localization of Dictyostelium homologues
of TRAP1 and GRP94 revealed by immunoelectron microscopy.
Exp Cell Res 303:415–424
Young JC, Moarefi I, Hartl FU (2001) Hsp90: a specialized but
essential protein-folding tool. J Cell Biol 154:267–273
Yu J, Nickels R, McIntosh L (2001) A genome approach to
mitochondrial–nuclear communication in Arabidopsis. Plant
Physiol Biochem 39: 345–353
Zhang Y, Dorey S, Swiderski M, Jones JDG (2004) Expression of
RPS4 in tobacco induces an AvrRps4-independent HR that
requires EDS1, SGT1 and HSP90. Plant J 40:213–224
Zhao Q, Wang J, Levichkin IV, Stasinopoulos S, Ryan MT,
Hoogenraad NJ (2002) A mitochondrial specific stress response
in mammalian cells. EMBO J 21:4411–4419
Zhao R, Davey M, Hsu Y-C, Kaplanek P, Tong A, Parsons AB,
Krogan N, Cagney G, Mai D, Greenblatt J, Boone C, Emili A,
Houry WA (2005) Navigating the chaperone network: an
integrative map of physical and genetic interactions mediated
by the Hsp90 chaperone. Cell 5:715–727
334 Plant Mol Biol (2008) 67:323–334
123