the plant journal (2005) 44 silencing a prohibitin alters...

Silencing a prohibitin alters plant development andsenescence

Jen-Chih Chen, Cai-Zhong Jiang and Michael S. Reid*

Department of Plant Sciences, University of California, Davis, One Shields Avenue, Davis, CA 95616, USA

Received 22 April 2005; revised 6 June 2005; accepted 16 June 2005.*For correspondence (fax þ530 752 4554; e-mail [email protected]).

Summary

Prohibitins, highly conserved mitochondrial proteins, have been shown to play important roles in cell cycling

and senescence in animals and yeast. Sequences with high similarity to prohibitins have been identified in a

number of plant species, but their function has not yet been demonstrated. The deduced amino acid sequences

of PhPHB1 and PhPHB2, sequences that we identified in a petunia floral expressed sequence tag (EST)

database, show high similarity to those of prohibitin-1 and prohibitin-2 proteins, respectively, reported from

yeast, animals and plants. Southern analysis suggested that these genes weremembers of small gene families

with at least two prohibitin-1 homologs and four prohibitin-2 homologs. When we downregulated expression

of prohibitin-1 using a Tobacco rattle virus-based (TRV), virus-induced gene silencing system (VIGS), we

observed plants with smaller and distorted leaves and flowers. Cells in silenced flowers were larger than in

control flowers, indicating a substantial reduction in the number of cell divisions that took place during corolla

development. The life of silenced flowers was shorter than that of controls, whether on the plant or detached.

The respiration of silenced flowers was higher than that of controls, and we observed a marked increase in the

abundance of transcripts of a catalase and a small heat-shock protein in the silenced flowers. Our data indicate

that prohibitins play a key role in plant development and senescence.

Keywords: flower, mitochondrion, programmed cell death, petunia, respiration, virus-induced gene silencing.

Introduction

Prohibitins are highly conserved proteins that have been

shown to play central roles in key processes, including cell-

cycle regulation, receptor-mediated signaling at the cell

surface, aging, apoptosis and mitochondrial function (Ber-

ger and Yaffe, 1998; Coates et al., 1997; McClung et al., 1992,

1995; Nijtmans et al., 2002; Piper et al., 2002). In animal

systems and yeast there are two classes of prohibitin,

exemplified by the yeast prohibitins Phb1p (prohibitin-1)

and Phb2p (prohibitin-2). The prohibitin-1 and prohibitin-2

polypeptides form a heterodimer that is the building block

for a barrel-shaped, high molecular-weight complex

(approximately 1.2 MDa) in the inner membrane of mito-

chondria in animals and yeast (Artal-Sanz et al., 2003;

Coates et al., 1997; Steglich et al., 1999). This complex ap-

pears to act as a chaperone, stabilizing components of the

respiratory chain (McClung et al., 1995; Nijtmans et al.,

2002; Steglich et al., 1999). There is a body of evidence

suggesting that prohibitins play a role in the regulation of

aging and senescence. Prohibitin expression declines

during senescence of both mammalian cells and yeast

(Coates et al., 1997; Piper et al., 2002), and overexpression of

prohibitins in mammalian cells increases cell viability fol-

lowing growth factor withdrawal (Vander Heiden et al.,

2002). In addition, mammalian tumor cell lines are charac-

terized by high prohibitin expression (Nijtmans et al., 2002).

Prohibitin-like sequences and proteins have been repor-

ted from maize, rice, tobacco and Arabidopsis (Nadimpalli

et al., 2000; Snedden and Fromm, 1997; Takahashi et al.,

2003). Although the plant prohibitins also appear to have a

mitochondrial location (Snedden and Fromm, 1997; Takaha-

shi et al., 2003), their biological function remains unknown.

Transgenic rice plants with altered prohibitin expression

were non-viable, suggesting that prohibitins may play a

critical role in plant growth and development analogous to

their role in other eukaryotes (Takahashi et al., 2003).

We are interested in the mechanisms of plant senescence

and cell death, and particularly in the control of floral

senescence, so we were intrigued by the possibility that

prohibitins may play, in plants, a role similar to that that has

been suggested for animals and yeast. Because attempts to

16 ª 2005 Blackwell Publishing Ltd

The Plant Journal (2005) 44, 16–24 doi: 10.1111/j.1365-313X.2005.02505.x

generate transgenic plants with down- or upregulated

expression of prohibitin were unsuccessful (Takahashi et al.,

2003), we decided to test the effect of downregulating

prohibitin expression using virus-induced gene silencing

(VIGS), a technique that avoids potentially embryo-lethal

effects. We report here the isolation, from petunia, of

sequences showing similarity to yeast Phb1p and Phb2p.

We also report the effects of downregulating prohibitin-1

expression in petunias using a Tobacco rattle virus (TRV)

silencing construct that employs chalcone synthase as a

reporter (Chen et al., 2004).

Results

Isolation of prohibitin-1 and prohibitin-2 homologs from

petunia

Using the BLAST search tool, we identified a putative pro-

hibitin, PhPHB1 (accession number CV294646), in a petunia

floral expressed sequence tag (EST) collection from the

University of Florida. The deduced amino acid sequence of

the full-length clone of PhPHB1 obtained by RACE-PCR

(accession number AY907015) shares 85% identity with a

tobacco prohibitin (Snedden and Fromm, 1997); 75%

identity with the Arabidopsis prohibitin-3 (accession

number AAD00157); 48% identity with budding yeast

Phb1p (accession number NP_011648); and 45% identity

with human prohibitin (accession number CAG46507)

(Figure 1).

Using primers designed to conserved regions in prohib-

itin-2 homologs reported from other solanaceous genera,

we isolated a partial prohibitin-2 sequence (PhPHB2, acces-

sion number DQ011266) from petunia floral cDNA. The

deduced amino acid sequence of this fragment (not shown)

shares 52% identity with the corresponding fragment of

PhPHB1; 90% identity with an Arabidopsis prohibitin-2

(AtPHB1, accession number AAN15530); 88% identity with

a rice putative prohibitin (OsPHB4, accession number

XP_477318); and 67% identity with budding yeast prohib-

itin-2 (Phb2p, accession number NP_011747.2). At the

Figure 1. Alignment of predicted amino acid

sequences of selected prohibitins.

Alignment of the predicted amino acid sequence

of a petunia prohibitin-1 (PhPHB1) with those of

prohibitins from Arabidopsis (AtPHB:

AAM64845.1); tobacco (NtPHB: AAC49690.1);

rice (OsPHB1: CAE76006); humans (hPHB:

AAS88903.1); and yeast (Phb1p: NP_011648.1,

Phb2p: NP_011747.2). Conserved amino acids

highlighted in black where identical; in gray

where similar.

Prohibitins in development and senescence 17

ª Blackwell Publishing Ltd, The Plant Journal, (2005), 44, 16–24

nucleotide level there was only 57% identity between the

sequences of PhPHB1 and PhPHB2.

Petunia prohibitins are encoded by multi-gene families

Southern analysis with probes for PhPHB1 and PhPHB2

showed that the petunia genome contains multiple genes

encoding prohibitin-1 and prohibitin-2 (Figure 2). At both

high and low stringency, two to four DNA fragments

hybridized with the PhPHB1 probe, and four to six fragments

hybridized with the PhPHB2 probe, depending on the

restriction enzyme used. The sizes of the fragments reacting

with the two probes were distinct.

Expression of PhPHB1 in different tissues and during floral

senescence

To assess the importance and possible role of the petunia

prohibitin, we examined its distribution in a range of tissues,

and during the opening and senescence of petunia flowers.

PhPHB1 transcripts were found atmoderate abundance in all

tested organs, including stems, flowers, leaves and roots

(data not shown). The abundance of PhPHB1 transcripts did

not change during flower opening, but declined after stage 3

as floral senescence progressed (Figure 3).

Virus-induced gene silencing of PhPHB1 in petunia

To examine the function of PhPHB1 in petunia, we cloned

different-sized fragments of PhPHB1 (Figure 4) into the TRV

CHS vector, and infected young petunia plants with Agro-

bacterium transformed with the tandem constructs. The

abundance of PhPHB1 transcripts in silenced plants was

Figure 2. Southern blot analysis of petunia

genomic DNA.

Genomic DNA was digested with different

restriction enzymes and membranes were

hybridized with PhPHB1 or PhPHB2 cDNA frag-

ments at high (left) and low (right) stringency.

S1

PhPHB1

18S

S2 S3 S4 S5

Figure 3. Abundance of prohibitin transcripts during development and sen-

escence of petunia flowers.

Total RNA was isolated from control flowers at the developmental stages

(1–5) described in Experimental procedures. The abundance of PhPHB1

transcripts in the flowers was visualized after 27 cycles of RT-PCR with

PhPHB1-specific primers. As a control, we carried out RT-PCR on all samples

with primers specific to 18S ribosomal RNA (15 cycles). Non-RT controls gave

no detectable signals (data not shown). The experiment was repeated with

two independent groups of flowers, with similar results.

Figure 4. Schematic representation of prohibitin-1 cDNA sequences from

petunia.

PhPHB1 includes the open reading frame and portions of the 5¢ and 3¢ UTR.

Fragments (i–iii) were inserted into the TRV-silencing vector in silencing

experiments. Arrows show location and direction of primers used to measure

transcript abundance. Fragment (i) was used as the probe in Southern blot

analysis.

18 Jen-Chih Chen et al.


strongly reduced (Figure 5), but the silencing had little effect

on PhPHB2 transcripts in the same plants. Real-time PCR

analysis of transcript abundance confirmed this observation:

PhPHB1 transcripts in silenced tissues were reduced to 12%

of the controls and PhPHB2 transcripts were unaffected

(data not shown).

Silencing of PhPHB1 had a strong impact on plant

development, generating phenotypes that were similar,

regardless of which PhPHB1 fragment we used for silencing,

and we used the 214 bp fragment in subsequent experi-

ments. All the silenced (TRV phb1/chs) plants (a minimum of

four plants were used for each construct and tests were

repeated at least three times) showed late flowering, a dwarf

phenotype and small, curling leaves (Figure 6a,b). Anther

development, as well as pollen formation, was also affected

in flowers from the TRV phb1/chs plants. Anthers in silenced

flowers were heart-shaped, white, and contained no mature

pollen grains (Figure 6c).

TRV phb1/chs flowers were also smaller than control

flowers (about half the diameter) and exhibited a range of

pale and purple pigmentation patterns (Figure 7). Unlike

infected flowers on TRV chs control plants, none of the

flowers or sectors on the TRV phb1/chs plants was com-

pletely white, suggesting that complete silencing might be

lethal.

Effects of silencing PhPHB1 on petal cell size and number

The smaller size of the pale flowers in TRV phb1/chs plants

suggested that silencing of prohibitin might result in smaller

and/or fewer cells in the petals. Light microscopic exam-

ination of epidermal peels showed that the cells in the pale

(silenced) sectors were considerably larger than those in the

purple (control) sectors (Figure 8), with a mean diameter 1.3

times that of cells in the control (purple) sectors or in white

sectors from TRV chs control plants.

Effect of silencing PhPHB1 on floral longevity.

The effect of silencing prohibitin expression on floral lon-

gevity was determined by tagging stage 2 silenced and

control flowers on a TRV phb1/chs plant and recording

the time for them to reach stage 5. Themean longevity of the

silenced flowers (7.3 � 1.6) was 5 days less than that of the

control flowers (12.5 � 1.0) (Figure 9).

Effect of silencing PhPHB1 on floral respiration

As prohibitins have been reported to be associated with the

mitochondrion (Snedden and Fromm, 1997; Takahashi et al.,

2003), we compared the rate and pattern of respiration of

control and silenced flowers from TRV phb1/chs plants. The

respiration rate of silenced flowers was significantly higher

than that of control flowers throughout flower aging. As

would be predicted from their shorter life on the plant, the

senescence-associated climacteric rise in respiration

occurred between 1 and 2 days earlier in the silenced flow-

ers (Figure 10).

Effect of silencing PhPHB1 on transcripts of genes encoding

ROS-scavenging and stress-related proteins

The earlier senescence and higher respiration of silenced

flowers from TRV phb1/chs plants suggested the possibility

that silencing prohibitin might result in mitochondrial mal-

function and the generation of reactive oxygen species

(ROS). To test this possibility, we examined the abundance

Figure 5. Virus-induced gene silencing of PhPHB1.

Transcript abundance in silenced and control tissues was determined using

RT-PCR and primers for PhPHB1, PhPHB2 and 18S ribosomal RNA. Products

were visualized after 30 cycles for the prohibitins and 15 cycles for the

ribosomal RNA. The experiment was repeated with three independent groups

of flowers, with similar results.

Figure 6. Pleiotropic effects of silencing PhPHB1 in petunia.

Growth habit (a), leaf morphology (b) and anthers (c) of control plants (left);

plants infected with Agrobacterium transformed with TRV carrying a

fragment of petunia CHS (center); or with TRV carrying fragments of CHS

and PhPHB1 in tandem (right). Photographs were taken 5 weeks after

infection.



of transcripts of a number of genes associated with electron

transport, with free radical and peroxide metabolism, and

with response to stress. The abundance of transcripts of

superoxide dismutase (MnSOD), ascorbate peroxidase and

four glutathione peroxidases was similar in silenced and

control flowers from TRV phb1/chs plants (Figure 11).

However, expression of a catalase (CAT) and of a small heat-

shock protein (sHSP) increased substantially in the silenced

flowers, while expression of one of two succinate dehy-

drogenase components (SDHip) fell slightly (Figure 11).

Quantification of transcript abundance using real-time PCR

showed a fourfold increase in CAT transcripts and a twofold

increase in sHSP transcripts in silenced flowers (data not

shown).

Discussion

Our data provide clear evidence that plant prohibitins,

known to be located in the mitochondrion (Snedden and

Fromm, 1997; Takahashi et al., 2003), play important roles in

cell cycling and aging, analogous to those already estab-

lished for other eukaryotes (Nijtmans et al., 2002). The

amino acid sequence deduced from the full-length sequence

of PhPHB1 shows high similarity with that of other prohib-

itin-1 sequences (Figure 1). It has been shown that at least

two isoforms exist in the prohibitin family (Nijtmans et al.,

2002). PhPHB1 and yeast prohibitin-1 (Phb1p) group in the

same cluster in a prohibitin phylogenetic tree (Figure 12),

indicating that PhPHB1 may be an ortholog of yeast Phb1p.

Figure 7. Effect of silencing PhPHB1 on petunia

flower morphology.

Petunias were infected with Agrobacterium

transformed with TRV carrying fragments of

CHS and PhPHB1 in tandem. Photographs of

unsilenced (left), uniformly silenced (center) and

partially silenced (right) flowers were taken

5 weeks after infection.

Figure 8. Effects of silencing PhPHB1 on petal

epidermal cell size.

Epidermal peels were removed from petals,

mounted in water and photographed using a

Zeiss compound microscope fitted with a pho-

tomicrography attachment. Peels were taken

from purple (a), pale (b) and mosaic (c) flowers

on a TRV phb1/chs plant, and from awhite flower

of a TRV chs plant (d).

Figure 9. Effect of silencing PhPHB1 on longevity of petunia flowers.

Petunia plants were infected with TRV PHB1/CHS. Control (left) and silenced

(right) flowers were photographed 5 days after anthesis (stage 2). The

photograph shows a typical example.

Figure 10. Effects of silencing PhPHB1 on respiration of petunia flowers.

Control (diamonds) and silenced (squares) flowers from TRV phb1/chs plants

were detached and their pedicels placed in water in small vials that were

sealed in glass jars ventilated with CO2-free air. The pattern of respiration

during floral senescencewas determined bymeasuring the CO2 content of the

exit air using an infrared gas analyzer, the output of which was recorded

digitally. The experiment was repeated four times; the graph shows a typical

comparison.



The phylogenetic tree shows that plant prohibitins separ-

ate into two families related to the two isoforms reported

from yeast and animal systems, prohibitin-1 and prohibitin-

2. The Arabidopsis genome contains seven prohibitin

sequences, three in the prohibitin-1 cluster and four in the

prohibitin-2 cluster. This analysis suggests that plant pro-

hibitins may play a similar role to those of Phb1p and Phb2p,

which form the membrane-associated supercomplex in

yeast.

The results of Southern analysis (Figure 2) indicate that

prohibitins are also a multigene family in petunia, which

raises the question as to whether the silencing phenotype

we observed is the result of reduced abundance of only

PhPHB1 transcripts. One of the advantages of VIGS is that

the use of conserved portions of the gene for silencing can

result in silencing of all members of the gene family. Given

the high similarity amongmembers of the prohibitin-1 clade

(Figure 1), we think it probable that the silencing phenotype

is the result of reduced transcript abundance of all petunia

prohibitin-1 species. The similarity at the nucleotide level

between the isolated petunia PhPHB1 and PhPHB2 se-

quences is relatively weak (57%), so silencing of one would

not be expected to affect the abundance of the other. Our

results confirm this expectation: levels of PhPHB2 tran-

scripts in the silenced and control tissues were similar.

Used with other target genes, the TRV chs silencing

system generates completely white flowers, indicating

strong silencing of the target gene (Chen et al., 2004). The

fact that we saw no such strong silencing in the TRV phb1/

chs plants suggests that complete silencing of PhPHB1 is

lethal, as was observed in attempts to silence a prohibitin

using antisense technology (Takahashi et al., 2003).

Reducing prohibitin-1 transcript abundance had a marked

effect in petunia plants, causing distortion of vegetative

organs, reducing organ size, and accelerating the onset of

floral senescence (Figures 6, 7 and 9). We hypothesized that

the smaller size of the petals of the TRV phb1/chs flowers

might be the result of fewer petal cells, reflecting impair-

ment of cell division, as has been reported for yeast, where

deletion of prohibitin genes resulted in a decrease in mean

replicative life span (the number of divisions available to

individual yeast cells) (Coates et al., 1997; Piper et al., 2002).

The fact that the cells in silenced petal sectors are larger,

even though the petals are smaller (Figures 7 and 8),

indicates a strong effect on cell division. As the silenced

flowers were approximately half the diameter of control

flowers (indicating approximately a quarter of the petal

area), and the cells had 1.3 times the diameter of the control

cells (which would give 1.7 times the projected area), we

Ctrl PHB

PHB

SDHip

SDHf

sHSP

CAT

MnSOD

APX

GPX1

GPX2

GPX2

GPX4

18S

Figure 11. Effects of silencing PhPHB1 on transcript abundance in petunia

flowers.

cDNAwas prepared from RNA extracted from corollas of control and silenced

flowers of plants that had been infected with TRV bearing PHB1/CHS. The

abundance of transcripts in flowers was visualized using 30 cycles of RT-PCR

with gene-specific primers. As a control RT-PCR was carried out on all

samples with primers specific to 18S ribosomal RNA (15 cycles). Non-RT

controls gave no detectable signals (data not shown). The experiment was

repeated with three independent groups of flowers, with similar results.

ZnPHB1OsPHB4

OsPHB3AtPHB2AtPHB1AtPHB6AtPHB5hBAP37Phb2pPhPHB1NtPHBAtPHB4AtPHB3

OsPHB2OsPHB1

AtPHB7Phb1phPHB

ZnPHB3

ZnPHB2

ZnPHB4

Figure 12. Phylogenetic tree for deduced amino acid sequences of reported

prohibitins, generated using PHYLIP (http://www.genebee.msu.su).



estimate the number of cells in the silenced petals to be 15%

of that in the control petals. During petal development,

therefore, the cells in the silenced petals underwent between

three and four fewer divisions than those in control petals,

supporting the hypothesis that plant prohibitins are required

for cell division. In human cell-culture lines, in contrast,

prohibitin proteins appear to play an anti-proliferation role

(McClung et al., 1992).

In animal systems, mitochondria play an important role in

the initiation of programmed cell death (PCD). The mitoch-

ondrial intermembrane space contains several cell-death

activators such as cytochrome c, and effectors such as

nucleases. In response to a variety of stimuli, amitochondrial

permeability transition (MPT), facilitated by a pore-mediated

permeability increase in the outer membrane, results in

release of these activators and effectors from the mitochon-

drion to trigger the caspases that are the primary step in the

cell-death pathway (Jiang and Wang, 2004). Impairing the

prohibitin complex results in increased generation of ROS in

yeast and Caenorhabditis elegans (Artal-Sanz et al., 2003;

Nijtmans et al., 2000) and ROS, in turn, have been shown to

trigger theMPT (Beckman andAmes, 1998; Crompton, 2004).

Recent evidence suggests that similar regulatory mecha-

nisms are involved in plant PCD. A role for mitochondria in

PCD has been documented in a number of plant systems.

Loss of mitochondrial membrane potential was found to be

an early marker in Arabidopsis PCD (Yao et al., 2004). MPT

has also been implicated in cell death induced by Victorin, a

host-selective toxin produced by Cochliobolus victoriae in

oats (Curtis and Wolpert, 2002). In addition, oxidative stress

leads to elevation of respiration and generation of ROS,

resulting in depletion of ATP production, alteration of MPT,

and PCD in Arabidopsis cells (Tiwari et al., 2002). Petal

senescence has many of the features of PCD (Eason et al.,

2002), but there has previously been no evidence of a role for

mitochondria in the process.

Wehypothesize that prohibitin-1 plays an important role in

mitochondrial homeostasis and control of the onset of floral

senescence. The N-terminal region of PhPHB1 has strong

similarity to the same region of the tobacco prohibitin that

has been shown to be located inmitochondria (Snedden and

Fromm, 1997). In other systems, it has been suggested that

the prohibitin complex forms a novel membrane-associated

holdase/unfoldase chaperone that stabilizes mitochondrial

proteins specifically required in situations of metabolic

stress (Nijtmans et al., 2000). It has also been demonstrated

that prohibitins regulate membrane protein degradation by

them-AAA protease inmitochondria (Steglich et al., 1999). It

is known that deficiencies in the respiratory chain complex

can lead to dysfunction of mitochondria and increased

production of ROS (Genova et al., 2004; Kayser et al., 2004).

If reducing levels of the prohibitin-1 proteins in petunia

petals compromised the stability of the respiratory chain

proteins, we would expect increased generation of ROS,

which might well result in accelerated senescence. The fact

that levels of respiration (Figure 10), transcripts encoding

catalase (a ROS-scavenging enzyme), and those encoding a

small heat-shock protein known to be a response to oxidative

stress conditionswere elevated in silenced flowers fromTRV

phb1/chs plants (Figure 11) supports the hypothesis that

prohibitin-1 is somehow involved in mitochondrial home-

ostasis. Additional supporting evidence is that silenced

flowers senesced more rapidly (Figure 9), and that the level

of PhPHB1 transcripts fell dramatically during the onset of

senescence in control petunia flowers (Figure 3).

Mitochondrial dysfunction may also explain the impair-

ment of cell division, anther development and other devel-

opmental processes (Figures 6–9) in silenced plants, as ATP

generation may be inadequate for the energy-intensive

processes of DNA replication, cell division and pollen

formation (Elorza et al., 2004; Hanson, 1991). We continue

to explore the role of the mitochondrion and the prohibitin

complex in flower development and senescence, as well as

in other aspects of plant growth, development and disease

responses.

Experimental procedures

Plant material

Petunia (Petunia · hybrida cv. Primetime Blue) seeds were gener-ously provided by Goldsmith Seeds (Gilroy, CA, USA). Plants weregrown in growth chambers at 22�C under 16 h light/8 h dark cycles.

Isolation and sequence analysis of petunia prohibitins

(PhPHB1 and PhPHB2)

A partial petunia prohibitin-1 (PhPHB1) sequence, which includedthe 3¢UTR region, was identified from a petunia floral EST database(University of Florida) and RACE-PCR was used to obtain the 5¢nucleotide sequence. A 437 bp partial petunia prohibitin 2 (PhPHB2)fragment was isolated from floral cDNA by RT-PCR using primers5¢-CACTTCGGGAAGTCGTGACCT-3¢ and 5¢-AATCAGCTGAGCACT-CTTAGC-3¢.

BLAST and CLUSTALX analysis tools were used to compare thenucleotide and deduced amino acid sequences of PhPHB1 andPhPHB2 with those of prohibitins reported from other organisms.A phylogenetic tree for selected prohibitins was generated usingPHYLIP (http://www.genebee.msu.su).

Southern blot analysis

Genomic DNA (5 lg per lane) was digested with various restrictionenzymes, separated on 0.7% agarose gel, transferred to a Hybond-Nþ membrane (Amersham Biosciences, Piscataway, NJ, USA), andhybridized to prohibitin cDNAs that were labeled using the AlkPhosDirect Labeling system (Amersham RPN 3680). Membranes werewashed at 50�C (low stringency) or 70�C (high stringency). Prohib-itin-1 detection was with an 841 bp PhPHB1 cDNA fragment (Fig-ure 4), and prohibitin-2 detection was with a 437 bp PhPHB2 cDNAfragment.



VIGS of PhPHB1

The pTRV1 and pTRV2 vectors for VIGS were kindly provided by DrDinesh-Kumar, Yale University, USA, and have been described indetail (Liu et al., 2002).

CHS construct. A 194 bp fragment of the CHS gene correspond-ing to bases 654–847 of petunia CHSJ (GenBank accession numberX14599) was PCR-amplified from petunia cDNA using primers 5¢-gctctagaACCATTGGGCATTTCTG-3¢with anXbaI restriction site and5¢-cggaattcAGCCTTTCTCATTTCATCC-3¢ with an EcoRI restrictionsite. The resulting product was cloned into pTRV2 to form pTRV2CHS which was used as a control in silencing experiments.

PhPHB1/CHS constructs. An 841 bp fragment of PhPHB1 wasPCR-amplified from petunia cDNA using primers 5¢-AAAATGGG-TAGCAACCAAGCAGCAGT-3¢ and 5¢-AACGCGAGGCATTGAGTCC-3¢. The 521 bp PhPHB1 cDNA fragment was the petunia EST cloneCV294646. A 214 bp fragment of the PhPHB1 gene was PCR-ampli-fied from petunia cDNA using primers 5¢-GGCTGCGATTATTA-GGGCTGAA-3¢ and 5¢-AACGCGAGGCATTGAGTCC-3¢.

These products were cloned into pTRV2-CHS to generate inde-pendent pTRV PHB1/CHS constructs. Four young petunia plantswere infected with Agrobacterium transformed with each of thesepTRV CHS/PHB1 constructs, or with pTRV2 CHS (control), asdescribed by Chen et al. (2004).

RT-PCR analysis

Total RNA was extracted from petunia tissues using TRIzol Reagent(Invitrogen, Carlsbad, CA, USA) and treatedwith RNase-free DNase I(Promega, Madison, WI, USA) to remove any contaminating ge-nomic DNA. First-strand cDNA, synthesized as described in detail byChen et al. (2004), was used as template for semi-quantitative andreal-time quantitative PCR. The PCR primers for amplifying VIGStarget gene transcripts were designed outside the region targetedfor gene silencing to avoid amplification from the TRV/VIGS gen-ome. The primers for detecting CHS were 5¢-ATGGCTCCTTCTCTT-GATG-3¢ and 5¢-AATCTTAGACTTGGGCTGGC-3¢. The primers 5¢-CATAAGCGCAACCGTCGTCAACTC-3¢ and 5¢-AATGTCGGGTAATC-GGGCAACTTC-3¢ were designed for PhPHB1. The primers forPhPHB2 were the same as those used in PhPHB2 cDNA isolation(above). The primers for SDHip (accession number CV297105) were5¢-TCTACGATCACACCATTGCCTCA-3¢ and 5¢-CATCACGGCTGTCC-ATAATCC-3¢. For SDHf (accession number CV292914), primers 5¢-GGCTTGTGCTAACAGGGTTGCT-3¢ and 5¢-CGCATGCATTGATTAG-CAGGTT-3¢ were used. Primers 5¢-AGCACCACCTCCAACATTCCAT-3¢ and 5¢-CCCTTTGCATTCTTAGGCAGTGA-3¢ were for a small heat-shock protein (accession number CV293761). Primers 5¢-TCA-CAGGCTGACAAGTCTCTCG-3¢ and 5¢-CCTTGGGAGCGAAACAAG-GTAA-3¢ were for CAT1 (accession number CV294789). Theabundance of 18S rRNA, determined using the amplification prim-ers 5¢-CATGGCCGTTCTTAGTTGGTGGAG-3¢ and 5¢-AAGAAGCTG-GCCGCGAAGGGATAC-3¢, served as an internal control.

Cell size

Cell size in control (purple) and silenced (white) sectors of flowersfrom TRV phb1/chs plants was determined by removing an epi-dermal peel from the petal, mounting it in water, and photograph-ing the cells using a Zeiss compound microscope fitted with aphotomicrography attachment (Zeiss, Jena, Germany). Cell diam-

eters were compared by measuring 20 replicate cells from silencedand control sectors of different flowers on the printed image.

Flower longevity

Determination of flower longevity was carried out as describedpreviously (Pech et al., 1987). The process of senescence in petuniaflowers was divided into five stages: (i) 1 day before opening; (ii)flowers at anthesis and fully open; (iii) incipient senescence (petalssoftening but corolla shape normal); (iv) mid-senescence (edges oftrumpet beginning to wilt); (v) late senescence (more than half thecorolla wilted).

The effect of PhPHB1 silencing on flower longevity was deter-mined by marking newly opened control and silenced flowers onthe same TRV phb1/chs plants, and following the progression ofsenescence for each marked flower. Three TRV phb1/chs plantswere used, and a total of 20 control and 20 silenced flowers wereexamined during the experiment.

Respiration

The pedicels of detached flowers were placed in water in small vialsthat were sealed in 65 ml glass jars ventilatedwith 0.2 L h)1 CO2-freeair and held at 20�C. The pattern of respiration (ml CO2 kg)1 FW h)1)during floral senescence was determined by measuring the CO2

content of the exit air using an infrared gas analyzer (QUBIT,Ontario, Canada), the output of which was recorded digitally(Imanishi et al., 1994).

Acknowledgements

We appreciate the assistance of Dr Dinesh-Kumar, Yale University,in providing the TRV constructs. We thank Dr David Clark, Universityof Florida, for access to the petunia EST database and EST clones; DrDavid Neale, Forest Genetics Institute, USFS, for use of the real-timequantitative PCR equipment; and Dr Richard Bostock, UC Davis, forassistance with Southern analysis. Goldsmith Seeds generouslydonated seeds of blue-flowered hybrid petunia cultivars. This studywas partially supported by a grant to M.S.R. from the AmericanFloral Endowment.

References

Artal-Sanz, M., Tsang, W., Willems, E., Grivell, L., Lemire, B., van

der Spek, H. and Nijtmans, L. (2003) The mitochondrial pro-hibitin complex is essential for embryonic viability and germ-line function in Caenorhabditis elegans. J. Biol. Chem. 278,32091–32099.

Beckman, K. and Ames, B. (1998) The free radical theory of agingmatures. Physiol. Rev. 78, 547–581.

Berger, K. and Yaffe, M. (1998) Prohibitin family members interactgenetically with mitochondrial inheritance components in Sac-charomyces cerevisiae. Mol. Cell Biol. 18, 4043–4052.

Chen, J.-C., Jiang, C.-Z., Gookin, T., Hunter, D., Clark, D. and Reid,

M. (2004) Chalcone synthase as a reporter in virus-induced genesilencing studies of flower senescence. Plant Mol. Biol. 55, 521–530.

Coates, P.J., Jamieson, D.J., Smart, K., Prescott, A.R. and Hall, P.A.

(1997) The prohibitin family of mitochondrial proteins regulatereplicative lifespan. Curr. Biol. 7, 607–610.

Crompton, M. (2004) Mitochondria and aging: a role for the per-meability transition? Aging Cell, 3, 3–6.



Curtis, M.J. and Wolpert, T.J. (2002) The oat mitochondrial per-meability transition and its implication in victorin binding andinduced cell death. Plant J. 29, 295–312.

Eason, J.R., Ryan, D.J., Pinkney, T.T. and O’Donoghue, E.M. (2002)Programmed cell death during flower senescence: isolation andcharacterization of cysteine proteinases from Sandersonia aur-antiaca. Funct. Plant Biol. 29, 1055–1064.

Elorza, A., Leon, G., Gomez, I., Mouras, A., Holuigue, L., Araya, A.

and Jordana, X. (2004) Nuclear SDH2-1 and SDH2-2 genes,encoding the iron–sulfur subunit of mitochondrial complex II inArabidopsis, have distinct cell-specific expression patterns andpromoter activities. Plant Physiol. 136, 4072–4087.

Genova, M., Pich, M., Bernacchia, A., Bianchi, C., Biondi, A., Bovina,

C., Falasca, A., Formiggini, G., Castelli, G. and Lenaz, G. (2004)The mitochondrial production of reactive oxygen species inrelation to aging and pathology. Ann. N Y Acad. Sci. 1011, 86–100.

Hanson, M. (1991) Plant mitochondrial mutations and male sterility.Annu. Rev. Genet. 25, 461–486.

Imanishi, H., Halevy, A., Kofranek, A., Han, S. and Reid, M. (1994)Respiratory and carbohydrate changes during ethylene-mediatedflower induction in Dutch iris. Sci. Hortic. 59, 275–284.

Jiang, X. and Wang, X. (2004) Cytochrome c-mediated apoptosis.Annu. Rev. Biochem. 73, 87–106.

Kayser, E., Sedensky, M. and Morgan, P. (2004) The effects ofcomplex I function and oxidative damage on lifespan andanaesthetic sensitivity in Caenorhabditis elegans. Mech. AgeingDev. 125, 455–464.

Liu, Y., Schiff, M. and Dinesh-Kumar, S.P. (2002) Virus-induced genesilencing in tomato. Plant J. 31, 777–786.

McClung, J., King, R., Walker, L., Danner, D., Nuell, M., Stewart, C.

and Dell’Orco, R. (1992) Expression of prohibitin, an antiprolifer-ative protein. Exp. Gerontol. 27, 413–417.

McClung, J.K., Jupe, E.R., Liu, X.T. and Dell’Orco, R.T. (1995) Pro-hibitin: potential role in senescence, development, and tumorsuppression. Exp. Gerontol. 30, 99–124.

Nadimpalli, R., Yalpani, N., Johal, G. and Simmons, C. (2000) Pro-hibitins, stomatins, and plant disease response genes compose aprotein superfamily that controls cell proliferation, ion channelregulation, and death. J. Biol. Chem. 275, 29579–29586.

Nijtmans, L., de Jong, L., Artal-Sanz, M., Coates, P., Berden, J.,

Back, J., Muijsers, A., van der Spek, H. and Grivell, L. (2000)Prohibitins act as a membrane-bound chaperone for the stabil-ization of mitochondrial proteins. EMBO J. 19, 2444–2451.

Nijtmans, L., Artal-Sanz, M., Grivell, L. and Coates, P. (2002) Themitochondrial PHB complex: roles in mitochondrial respiratorycomplex assembly, ageing and degenerative disease. Cell. Mol.Life Sci. 59, 143–155.

Pech, J.C., Latche, A., Larrigaudiere, C. and Reid, M.S. (1987) Controlof early ethylene synthesis in pollinated petunia flowers. PlantPhysiol. Biochem. 25, 431–437.

Piper, P., Jones, G., Bringloe, D., Harris, N., MacLean, M. and

Mollapour, M. (2002) The shortened replicative life span of pro-hibitin mutants of yeast appears to be due to defective mitoch-ondrial segregation in old mother cells. Aging Cell, 1, 149–157.

Snedden, W. and Fromm, H. (1997) Characterization of the planthomologue of prohibitin, a gene associated with antiproliferativeactivity in mammalian cells. Plant Mol. Biol. 33, 753–756.

Steglich, G., Neupert, W. and Langer, T. (1999) Prohibitins regulatemembrane protein degradation by the m-AAA protease in mito-chondria. Mol. Cell Biol. 19, 3435–3442.

Takahashi, A., Kawasaki, T., Wong, H.L., Suharsono, U., Hirano, H.

and Shimamoto, K. (2003) Hyperphosphorylation of a mitoch-ondrial protein, prohibitin, is induced by calyculin A in a ricelesion-mimic mutant cdr1. Plant Physiol. 132, 1861–1869.

Tiwari, B., Belenghi, B. and Levine, A. (2002) Oxidative stressincreased respiration and generation of reactive oxygen species,resulting in ATP depletion, opening of mitochondrial permeabil-ity transition, and programmed cell death. Plant Physiol. 128,1271–1281.

Vander Heiden, M., Choy, J., Vander Weele, D., Brace, J., Harris, M.,

Bauer, D., Prange, B., Kron, S., Thompson, C. and Rudin, C. (2002)Bcl-x(L) complements Saccharomyces cerevisiae genes thatfacilitate the switch from glycolytic to oxidative metabolism.J. Biol. Chem. 277, 44870–44876.

Yao, N., Eisfelder, B.J., Marvin, J. and Greenberg, J.T. (2004) Themitochondrion – an organelle commonly involved in pro-grammed cell death in Arabidopsis thaliana. Plant J. 40, 596–610.



the plant journal (2005) 44 silencing a prohibitin alters...

Documents