Báo cáo khoa học: Phytoene synthase genes in tomato (Solanum lycopersicum L.) – new data on the structures, the deduced amino acid sequences and the expression patterns doc

In the present study, we report on the genomic structures of the Psy1 and Psy2 genes and on their transcription patterns in different tomato tissues.. Oligonucleotide primer and probe se

lycopersicum L.) – new data on the structures, the deduced amino acid sequences and the expression patterns

Giovanni Giorio, Adriana Lucia Stigliani and Caterina D’Ambrosio

Metapontum Agrobios, Metaponto, Italy

Fruits are the mechanism by which angiosperms

dis-perse seeds and are the result of a tight co-evolution

between plants and their seed dispersers [1] Tomato

(Solanum lycopersicum L.) belongs to the Solanaceae

(Nightshade) family, which contains many

differenti-ated taxa occurring worldwide Its fruit type is the

berry: red, fleshy and with a pulpy interior rich in seeds [2] Among the 12 wild relatives of tomato, there

is only one (Solanum pimpinellifolium B Juss.) with a red berry and two with yellow, yellow–green or orange fruits [Solanum cheesmaniae (L Riley) Fosberg; Sola-num galapagenseS.C Darwin and Peralta], whereas all

Keywords

carotenoid metabolism; chloroplast;

chromoplast; fruit colour; phytoene synthase

Correspondence

G Giorio, Metapontum Agrobios, SS Jonica

Km 448.2, Metaponto, MT 75010, Italy

Fax: +39 0835 740204

Tel: +39 0835 740276

E-mail: ggiorio@agrobios.it

All authors contributed equally to this work

(Received 22 October 2007, revised 27

November 2007, accepted 3 December

2007)

doi:10.1111/j.1742-4658.2007.06219.x

The fruit of tomato (Solanum lycopersicum L.) is a berry: red, fleshy and rich in seeds Its colour is due to the high content of lycopene whose syn-thesis is activated by the phytoene synthase 1 (PSY1) enzyme, encoded by Psy1 which is distinct from Psy2 In the present study, we report on the genomic structures of the Psy1 and Psy2 genes and on their transcription patterns in different tomato tissues Our results have completely clarified the structure of the Psy1 and Psy2 genes in the coding sequence region The two genes were shown to have an highly conserved structure, with seven exons being almost identical and six introns being much more vari-able For Psy1 and Psy2, respectively, the sequenced regions were 4527 and

3542 bp long, the coding sequences were 1239 bp and 1317 bp long, whereas the predicted protein sequences were 412 and 438 amino acids The two proteins are almost identical in the central region, whereas most differences are present in the N-terminus and C-terminus Quantitative real time PCR analysis showed that Psy2 transcript was present in all tested plant tissues, whereas Psy1 transcript could be detected in chromoplast-containing tissues, particularly in fruit where it activates and boosts lyco-pene accumulation Interestingly, the organ with the highest relative content

of Psy2 transcript is the petal and not the leaf Psy1 is a Psy2 paralog derived through a gene duplication event that have involved other genes encoding rate controlling enzymes of the carotenoid pathway Duplicate genes have been recruited to allow carotenoid synthesis in petals and fruits However, recruitment of carotenoid metabolism for fruit pigmentation could have occurred later in the evolution, either because phytoene syn-thase gene duplication occurred later or because the fruit pigmentation pro-cess required a more sophisticated mechanism involving tight control of the transcription of other genes

Abbreviations

cTP, chloroplastic transit peptide; PSY, phytoene synthase; qRT-PCR, quantitative RT-PCR.

the other species have green, yellow–green, dark green

or black fruits [3,4]

During the development of tomato fruit, the shift

from green to red colour is due to the degradation of

chlorophylls and the accumulation of the carotenoid

lycopene

Carotenoid pathways in plants have been described

in great detail using genetic, biochemical and

mole-cular data, mainly from Arabidopsis [5,6]

The first step in the synthesis of lycopene is the

condensation of two molecules of geranylgeranyl

diphosphate to form the 15-cis-isomer of phytoene

This two-step reaction is catalysed by the enzyme

phytoene synthase (PSY) Following four desaturations

and probably two [7] isomerization steps, the 15-cis

phytoene is converted to all-trans lycopene by

phyto-ene desaturase, f-carotphyto-ene desaturase and carotphyto-ene

isomerases At this point, the pathway is branched

because the lycopene can be converted to lutein, which

appears to be the end-product of the first branch, or

to zeaxanthin, which can be further converted to

violaxanthin In plants, carotenoids are mainly

involved in photosynthesis as accessory pigments, in

photoprotection (quenching and xanthophylls cycle)

and in the formation of abscisic acid Moreover, many

species use them to make coloured flowers and fruits

to attract pollinators and seed dispersers In tomato,

for example, the flower has a bright yellow–orange

corolla resulting from the combined effect of

chromo-phores of neoxanthin, violaxanthin and lutein [8],

whereas its fruit, owing to the high content of

lycopene, is of a deep red colour at the end of the

developmental process However, in tomato, two active

forms of PSY have been described, PSY1 and PSY2,

each encoded by its own gene, Psy1 and Psy2

Although the first report on the cDNA sequence of

Psy1gene (pTOM5) in tomato (X60440) can be traced

back to Ray et al [9], the last annotation of the

derived protein (P08196) still reports the presence of

conflicting data Moreover, the Psy2 cDNA sequence

available in the NCBI database (L23424) is incomplete

because it is devoid of the 5¢-region coding for the

pro-tein amino-terminus The correct DNA and proteic

sequences of PSY1 were first reported by Bartley et al

[10], who demonstrated the correctness of the

hypothe-sis of Armstrong et al [11], which proposed that

pTOM5 may encode the tomato homolog of the

bifunctional red pepper PSY [12] As for the Psy2

gene, the first report on the DNA sequence was

pro-vided by Bartley and Scolnik [13] These authors

dem-onstrated that the genomic clone F (X60440), which

was considered to be a PSY1 pseudogene by Ray et al

[14], was indeed the DNA sequence of the paralogous

PSY1 gene coding for a PSY enzyme active in photo-synthetic tissues

In the present study, we report on the genomic structures of Psy1 and Psy2 genes and the transcrip-tion patterns of both genes in different tomato tissues The role of carotenoids as secondary metabolites in the pigmentation of tomato flowers and fruits has been also reanalysed in the light of these results

Results

Isolation of tomato Psy1 and Psy2 genes Using the sequences M8474 and L23424 reported in the NCBI database corresponding, respectively, to tomato Psy1 and Psy2 mRNAs, an extended database search-ing ussearch-ing blast program was conducted aimsearch-ing to reconstruct the entire coding sequences of the two genes

The need for a reconstruction was based on the lack

of information regarding the 5¢-region in the Psy2 DNA sequence and from the conflicting evidence avail-able in the database records of the PSY1 gene Using the reconstructed sequences, a set of primers was designed and used to amplify the cDNAs derived from fruit or leaf RNAs of tomato cultivar Red Setter (Table 1) The Psy1 and Psy2 cDNAs were cloned in suitable vectors, sequenced and deposited in the NCBI database as EF534739 and EF534738 Combining these two mRNA sequences with those of GTOM5 (X60441) and clone F (X60440), two sets of primers were designed to amplify genomic DNA fragments corresponding to the introns of the two genes The fragments were sequenced and enabled the complete reconstruction of the Psy1 and Psy2 genes with the annotation of introns and exons The GenBank acces-sion numbers of the two genes are EF534740 (Psy1) and EU021055 (Psy2) However, the UTR regions for both genes were only partially reconstructed

The comparison of the two genes (Fig 1 and Table 2) showed a strong conservation of the gene structure The sequenced regions of the two genes were

4527 and 3542 bp long, respectively, for Psy1 and Psy2 The two genes contain at least seven exons and six introns The data are not conclusive because the UTR regions were only partially sequenced The cod-ing sequences were 1239 bp for Psy1 and 1317 bp for Psy2, with the latter being longer in the 5¢-region The start codon is located in the second exon in both genes Intron and exon numeration was changed com-pared to the original annotation of the Psy1 gene based on the GTOM5 sequence (X60441) because we discovered an additional intron upstream of the start

of the GTOM5 clone Length of the first intron is

942 bp in Psy1 and 153 bp in Psy2 Corresponding

exons in the central part of the two genes (i.e exons

3, 4, 5 and 6) have the same length Exons 2 and 7

have different length between the two genes because

they code for the regions in which the two proteins

are different Exon 1 shows the greatest difference because it was only partially sequenced in both genes and, therefore, the start of characterization was dif-ferent in the two genes As expected, a comparison of intron lengths between the two genes revealed great variability

Table 1 Oligonucleotide primer and probe sequences used for cloning and for transcript quantitation using qRT-PCR analysis of Psy1 and Psy2 genes.

Gene Primer or probe name Primer nucleotide sequences (5¢- to 3¢)

GenBank accession number

Amplicon

Le18SrRNA-R-179 CCCCGTGTTAGGATTGGGT

TaqMan-Le18s-140 AAGGCAGCAGGCGCGCAAA

Fig 1 Comparison of the structures of the Psy1 and Psy2 genes.

Table 2 Structure of tomato Psy1 and Psy2 genes.

Gene

Length 5¢-UTR CDS 3¢-UTR Length

Integral cTPb Mature

Amino acids kDa

Amino acids kDa Psy1 32 656 51 173 236 193 181 942 120 423 313 518 689 4527 276 1239 7 1522 412 45.32 62 38.50 Psy2 112 710 51 173 236 193 199 153 107 710 273 227 398 3542 338 1317 19 1674 438 48.18 86 38.72

a Exons 1 and 7 were only partially sequenced as well as the 5¢- and 3¢-UTRs b Predicted by the TARGETP

The predicted protein sequences were 412 and 438

amino acids for PSY1 and PSY2, respectively

Align-ment of the two protein sequences (Fig 2) showed

78% residue identity (341⁄ 438) The central part of

two proteins are similar, whereas the major differences

are present in the N-terminus In particular, both

protein sequences start with the sequence

MSVALLWVVSP and the PSY2 protein has two

sequences of four and 19 residues that are not present

in the PSY1 sequence Moreover, the last four

resi-dues, SLQR, at C-terminus of PSY1 are replaced by

the sequence SPLAKT in PSY2 targetp and

predo-tar software were used to predict the presence of

putative N-terminal targeting sequences [15,16] Both

proteins were predicted by predotar to have a plastid

target signal Conversely, targetp predicted a

chloro-plastic transit peptide (cTP) of 62 amino acids only for

the PSY1 protein In this case, a mature PSY1 in the

plastids would have a predicted size of 38.5 kDa,

which is agreement with previous experimental

evi-dence [17] targetp failed to predict a subcellular

localization for PSY2 when the entire sequence was

submitted However, when the query sequence

con-tained residues 1 to 91–95 of PSY2 protein, the

soft-ware always detected a chloroplastic transit peptide of

86 amino acids This may be due to the presence of

specific motifs beyond the first 95 residues that

inter-fere with the prediction

Transcription analysis in tomato tissues

Psy1 and Psy2 transcript contents were estimated in

RNA samples derived from root, leaf, petal, anther,

ovary and fruit at three developmental stages (Mature Green, Pink and Ripe) using quantitative RT-PCR (qRT-PCR) with gene-specific fluorescent probes (Fig 3)

Since the estimates of Psy1 and Psy2 relative tran-script contents were normalized onto the 18S rRNA (endogenous reference) transcript contents and com-pared to normalised petal transcript content (calibra-tor), it is possible for each gene to make an easy comparison of the relative transcript contents among the nine tissues (Figs 4 and 5) Psy1 transcript was absent in root RNA, whereas it could be detected in leaf, sepal, ovary and in the fruit at mature green stage However, in these tissues, Psy1 transcript con-tent ranged between 2% and 3% of the concon-tent in the petal This organ appeared to contain a considerable amount of Psy1 transcript As expected, Psy1 tran-script showed a typical increase between the Mature Green and Pink and a reduction between the Pink and Ripe stages

The Psy2 transcription pattern was quite unpredict-able in that the organ with the highest content was shown to be the petal instead of the leaf where the PSY2 enzyme has a pivotal role in the assembly of photosynthetic apparatus Very similar amounts of this transcript were detected in leaf, sepal and ovary RNAs Fruit RNAs contained Psy2 transcript at a level allowing easy detection, although it was very low compared to the content in petal RNA By contrast to Psy1, the Pys2 transcript was detectable in root RNA Estimates of the relative contents of the two tran-scripts for each tissue were derived with the compara-tive Ct methods The differences between the estimates Fig 2 Sequence alignment of PSY1 and PSY2 protein derived from analysis with CLUSTAL W program (EMBL).

A B

C

D

Fig 3 Organs of tomato assayed for Psy1

and Psy2 transcript content (A) Roots (B)

Expanding leaf (C) Flower organs: calyx

(sepals), corolla (petals), stamen cone

(sta-mens) and pistil (ovary, style and stigma).

(D) Fruit developmental stages: MG, Mature

Green; P, Pink; R, Ripe.

Fig 4 Transcription analysis of tomato Psy1 gene carried out using

qRT-PCR with gene-specific fluorescent probes on transcripts from

nine different tissues At least two RNA samples were assayed for

each tissue Three replicated reactions were performed for each

sample, both in the construction of standard curve and in the

quanti-tation of samples The estimates are expressed as the mean ± SD.

Fig 5 Transcription analysis of tomato Psy2 gene carried out using qRT-PCR with gene-specific fluorescent probes on transcripts from nine different tissues For details, see Fig 4.

of the threshold cycle means of the two genes for each

tissue were used as the exponent in the formula 2(DCt)

Accordingly, and assuming that the efficiency of the

amplification of the two genes was equal, it is possible

to obtain an estimate for each tissue of the transcript

content of Psy1 relative to Psy2 (Fig 6) In green

tis-sues (i.e in leaf, sepal and ovary), the content of Psy1

transcript was 0.4-fold lower than that of Psy2

tran-script Conversely, in pigmented tissues, such as petal,

anther and fruit at Pink and Ripe stages, the content

of Psy1 was shown to be much greater than that of

Psy2transcript The estimates ranged between 5.2-fold

greater in the petal to 213.3-fold greater in the fruit at

the Pink stage

Discussion

In the present study, we report on the molecular

char-acterization of the tomato Psy1 and Psy2 genes and

also provide the deduced complete amino acid

sequences of the two enzymes, together with new

insights into the transcriptional regulation of Psy1 and

Psy2 in chloroplast- and chromoplast-containing

tis-sues of tomato

The results obtained have completely clarified the

structure of the Psy1 and Psy2 genes in the coding

sequence region The two genes were shown to have a

highly conserved structure, with seven exons being

almost identical and six introns being much more

vari-able in sequence and length The two deduced protein

sequences showed high similarity in the central part

(86% residue identity) beyond the putative transit

pep-tide cleavage site The two putative cTPs, as predicted

by targetp, showed a greater diversity, with the PSY1

having two putative deletions of 4 and 19 residues Interestingly, the two proteins start with the sequence MSVALLWVVSP, which is the longest stretch of iden-tical residues in the N-terminus A blast search (blastp algorithm) of the NCBI protein database using this sequence as a query resulted in the retrieval of all PSY protein sequences belonging to dicotyledons Moreover, a search performed with the motif MSXXXXWVVXP was also able to retrieve all these proteins

With respect to chloroplastic transit peptide func-tion, the localization of PSY1 and PSY2 into the sub-compartments of plastids has not yet been com-pletely clarified, although extensive studies have been carried out in tomato [13,18,19] and in Narcissus [20–22] The results obtained in tomato were not conclusive, probably because of a confounding effect due to the two different forms of PSY However, as noted by Gallagher et al [23] in grass PSYs, differ-ences in the N-terminus as well as the C-terminus of the two proteins may result in differences in their plastid localization

Transcription analysis of the two genes using qRT-PCR clearly showed that, with the exception of Psy1

in the roots, transcripts of both genes are detectable in all tested tomato tissues Unexpectedly, the organ with the highest relative content of Psy2 transcript is the petal and not the leaf Psy2 transcript content in the leaf is only approximately 25% of that in the petal This result could not be anticipated because Psy2 was thought to be the chloroplast-specific PSY and no pre-vious report had addressed gene expression in this organ using a method as sufficiently sensitive as quan-titative real time PCR The high content of Psy2 tran-script in tomato petals could also explain why the flowers of yellow flesh mutants, r and ry, are pale or normal, respectively, whereas, in the lines in which the Psy1-derived transgene triggered a cosuppression of PSY genes, the flowers were almost white [24]

In the fruit, Psy2 transcript is detectable at all tested stages and appears to increase during ripening Finally, PSY2 must have some specific activities in the roots because Psy2 transcript is present in this tissue in con-trast to Psy1 transcript which is undetectable

Psy1 transcription analysis results were in accor-dance with those obtained in previous investigations [25,26] Psy1 transcript is almost undetectable in the fruit from the onset of maturation until the Mature Green stage From the Breaker stage onward, the tran-script level increases dramatically reaching its maxi-mum at the Pink stage and decreasing slowing with the progression of fruit ripening Its content in the fruit at Pink stage is almost three-fold greater than Fig 6 Transcript content of Psy1 relative to Psy2 across all

tissues.

that in the petals Nevertheless, Psy1 transcript content

in this organ was estimated to be 5.2-fold greater than

that of Psy2 which has its greatest expression in the

petal

Taken together, these results confirm the specialized

role of PSY1 in the colouring of fruit and that of

PSY2 in the synthesis of carotenoids involved in

pho-tosynthesis, in photoprotection (quenching and

xan-thophylls cycle) and in the formation of abscisic acid

Both genes appear, however, to be involved in the

flower colour process because their transcript contents

in the petal were very high, particularly Psy2

tran-script

Psy1is a Psy2 paralog derived through a gene

dupli-cation event After duplidupli-cation, the two genes have

been maintained in the genome owing to

subfunction-alization, which, in this case, is in the form of a

division of gene expression [27] The recruitment of

primary carotenoid metabolism as secondary

metabo-lism has been described in maize [23] as well as in

tomato for flower and fruit pigmentation [8] However,

in tomato, it has been hypothesised that recruitment

required duplication of all genes encoding the

rate-controlling enzymes of the pathway, namely carotene

beta-hydroxylase, lycopene cyclase and PSY, and that

the duplicated pathway was exploited originally for

flower pigmentation and only later for fruit

pigmenta-tion [8] The latter hypothesis serves to explain why all

13 tomato species have yellow coloured flowers,

whereas only three have red, yellow, yellow green or

orange coloured fruits [4] However, it is not known

whether the recruitment of the metabolism for fruit

pigmentation has occurred on a second occasion

because the PSY gene ancestor duplicated later or

because the subfunctionalization of the two paralogs

was more complex, thus requiring more time

By comparing the protein sequences of the PSY

par-alogs and that of carotene beta-hydroxylase parpar-alogs

(CrtR-b1, CAB55625; CrtR-b2, ABI23730), it is found

that a reduced similarity (73.7% residue identity and

82.6% similarity) is seen for CRTR-Bs compared to

PSYs (77.8% residue identity and 85.3% similarity)

that could indicate an early duplication of the CrtR-b

gene ancestor However, the recruitment of PSY1 for

fruit carotenoid metabolism has likely required a more

sophisticated mechanism involving the tight and timely

control of Psy1, Lcy-b and Lcy-e transcription during

fruit development Accumulation of lycopene in

tomato fruit, and therefore the colour shift from green

to red, starts at the Mature Green stage when the

seeds have completed their development and are able

to give rise to new plants This mechanism can be

con-sidered as a sort of light switch because the tomato

fruit is switched on at the appropriate time to appear like a red light in the green background of the plant canopy, thus alerting the seed dispersers

Experimental procedures

Plant materials and nucleic acid extraction Tomato plants (cv Red Setter) were grown in a green-house Total genomic DNA and total RNA were extracted

by leaf and fruit tissue samples using standard protocols Complementary DNA was synthesized from 1.5 lg of RNA using the ThermoScript RT-PCR System kit (Invi-trogen, Carlsbad, CA, USA) with random hexamer primers following the manufacturer’s instructions

Primer design, amplification of genomic DNA and cDNA and sequencing

Using the reconstructed sequences of Psy1 and Psy2 cDNAs, a set of primers were designed and used to amplify the cDNAs synthesized from fruit or leaf RNAs of the tomato cultivar Red Setter

Amplification was performed with the kit Phusion High-Fidelity DNA Polymerase (Finnzymes Oy, Espoo, Finland) in a PTC-200 thermal cycler (MJ Research, Bio-Rad Laboratories Inc., Hercules, CA, USA) in 20 lL reac-tion volume After checking the specificity of the reacreac-tions

by agarose gel electrophoresis analysis, an aliquot of the reaction was used to produce recombinant vectors with pCR-BLUNT II-TOPO (Invitrogen), which were trans-formed into competent Escherichia coli cells Plasmid DNA harbouring the two genes were isolated from recombinant cells and used for sequence analysis

Sequencing reactions were performed with the ABI PRISM BigDye Terminator v3.1 Cycle Sequencing kit and analysed with the Applied Biosystems 3130 Genetic Analyzer (Applied Biosystems, Foster City, CA, USA) Using specific software, the Psy1 and Psy2 partial mRNA sequences were assembled

Combining these two mRNA sequences with those of GTOM5 (X60441) and clone F (X60440), two sets of prim-ers were designed to amplify genomic DNA fragments cor-responding to the introns of the two genes After PCR, the amplified fragments were gel purified and sequenced using the protocols reported above

Quantitative analysis of Psy1 and Psy2 transcript contents (qRT-PCR)

Transcription analysis of tomato Psy1 and Psy2 genes was carried out using qRT-PCR with gene-specific fluorescent probes Transcript contents were estimated in RNA sam-ples derived from root, leaf, petal, anther, ovary and fruit

at three developmental stages (Mature Green, Pink and

Ripe)

Reactions were conducted in 96-well reaction plates in a

25 lL volume containing 12.5 lL of the Platinum

Quanti-tative PCR SuperMix-UDG (Invitrogen), 300 nm forward

primer, 300 nm reverse primer and 150 nm TaqMan probe

One microlitre of the cDNA sample (75 ng of RNA) was

used for each reaction qRT-PCR was performed using the

iCycler iQ Real Time PCR Detection System (Bio-Rad

Laboratories Inc.) Thermal cycling condition were 95C

for 3 min for activation of DNA polymerase, and 40 cycles

of 95C for 15 s and 60 C for 1 min Estimates of

tran-script content were derived using the standard curve

method [28] performing reactions in separate tubes

Stan-dard curves were prepared for both the target transcripts,

Psy1 and Psy2, and the endogenous reference 18S rRNA

gene using a petal cDNA stock sample, by assembling a set

of reactions using three-fold serial dilutions with six points

for 18S rRNA and eight points for both Psy1 and Psy2

Each PCR reaction was performed in triplicate, both for

the construction of the standard curves and for sample

quantitations Gene starting quantities for each sample

were estimated using regression parameter estimates of the

standard curve Estimates of Psy1 and Psy2 relative

tran-script contents were normalized onto the endogenous

refer-ence transcript (18S rRNA) to account for differrefer-ences in

the amount of total RNA content among samples and

com-pared with the normalised transcript content of petal, which

was chosen as calibrator Sequences of primers and

Taq-Man probes were the same as those previously used [25]

and are reported in Table 1

Acknowledgements

We wish to thank all colleagues of Metapontum

Ag-robios who collaborated in the project We are grateful

to Professor Peter Beyer (Freiburg, Germany) for

valu-able comments and helpful suggestions and Professor

Gerhard Sandmann (Frankfurt, Germany) for critical

reading of the manuscript

References

1 Knapp S (2002) Tobacco to tomatoes: a phylogenetic

perspective on fruit diversity in the Solanaceae J Exp

Bot 53, 2001–2022

2 Knapp S, Bohs L, Nee M & Spooner DM (2004)

Solan-aceae – a model for linking genomics with biodiversity

Comp Funct Genomics 5, 285–291

3 Peralta IE, Knapp S & Spooner DM (2006)

Nomencla-ture for wild and cultivated tomatoes Rep Tomato

Genet Coop 56, 6–12

4 Peralta IE & Spooner DM (2007) History, origin and

early cultivation of tomato (Solanaceae) In Genetic

Improvement of Solanaceous Crops, Vol 2: Tomato (Razdan MK & Mattoo AK, eds), pp 1–27 Science Publishers, Enfield, CT

5 DellaPenna D & Pogson BJ (2006) Vitamin synthesis in plants: tocopherols and carotenoids Annu Rev Plant Biol 57, 711–738

6 Sandmann G, Ro¨mer S & Fraser PD (2006) Under-standing carotenoid metabolism as a necessity for genetic engineering of crop plants Metab Eng 8, 291–302

7 Li F, Murillo C & Wurtzel ET (2007) Maize Y9 encodes a product essential for 15-cis-f-carotene isomer-ization Plant Physiol 144, 1181–1189

8 Galpaz N, Ronen G, Khalfa Z, Zamir D & Hirschberg

J (2006) A chromoplast-specific carotenoid biosynthesis pathway is revealed by cloning of the tomato white-flowerlocus Plant Cell 18, 1947–1960

9 Ray J, Bird C, Maunders M, Grierson D & Schuch W (1987) Sequence of pTOM5, a ripening related cDNA from tomato Nucleic Acids Res 15, 10587

10 Bartley GE, Viitanen PV, Bacot KO & Scolnik PA (1992) A tomato gene expressed during fruit ripening encodes an enzyme of the carotenoid biosynthesis path-way J Biol Chem 267, 5036–5039

11 Armstrong GA, Alberti M & Hearst JE (1990) Con-served enzymes mediate the early reactions of caroten-oid biosynthesis in nonphotosynthetic and

photosynthetic prokaryotes Proc Natl Acad Sci USA

87, 9975–9979

12 Dogbo O, Laferriere A, d’Harlingue A & Camara B (1988) Carotenoid biosynthesis: isolation and character-ization of a bifunctional enzyme catalyzing the synthesis

of phytoene Proc Natl Acad Sci USA 85, 7054–7058

13 Bartley GE & Scolnik PA (1993) cDNA cloning, expres-sion during development, and genome mapping of PSY2, a second tomato gene encoding phytoene syn-thase J Biol Chem 268, 25718–25721

14 Ray J, Moureau P, Bird C, Bird A, Grierson D, Maun-ders M, Truesdale M, Bramley P & Schuch W (1992) Cloning and characterization of a gene involved in phy-toene synthesis from tomato Plant Mol Biol 19, 401– 404

15 Emanuelsson O, Brunak S, von Heijne G & Nielsen H (2007) Locating proteins in the cell using TargetP, SignalP and related tools Nat Protoc 2, 953–971

16 Small I, Peeters N, Legeai F & Lurin C (2004) Predo-tar: a tool for rapidly screening proteomes for N-termi-nal targeting sequences Proteomics 4, 1581–1590

17 Fraser PD, Truesdale MR, Bird CR, Schuch W & Bramley PM (1994) Carotenoid biosynthesis during tomato fruit development (evidence for tissue-specific gene expression) Plant Physiol 105, 405–413

18 Fraser PD, Kiano J, Truesdale M, Schuch W & Bram-ley PM (1999) Phytoene synthase-2 enzyme activity in

tomato does not contribute to carotenoid synthesis in

ripening fruit Plant Mol Biol 40, 687–698

19 Fraser PD, Schuch W & Bramley PM (2000) Phytoene

synthase from tomato (Lycopersicon esculentum)

chlo-roplasts – partial purification and biochemical

proper-ties Planta 211, 361–369

20 Kreuz K, Beyer P & Kleinig H (1982) The site of

carotenogenic enzymes in chromoplasts from Narcissus

pseudonarcissusL Planta 154, 66–69

21 Schledz M, Al-Babili S, Lintig JV, Haubruck H,

Rab-bani S, Kleinig H & Beyer P (1996) Phytoene synthase

from Narcissus pseudonarcissus: functional expression,

galactolipid requirement, topological distribution in

chromoplasts and induction during flowering Plant J

10, 781–792

22 Bonk M, Hoffmann B, Lintig J, Schledz M, Al-Babili

S, Hobeika E, Kleinig H & Beyer P (1997)

Chloro-plast import of four carotenoid biosynthetic enzymes

in vitroreveals differential fates prior to membrane

binding and oligomeric assembly Eur J Biochem 247,

942–950

23 Gallagher CE, Matthews PD, Li F & Wurtzel ET (2004) Gene duplication in the carotenoid biosynthetic pathway preceded evolution of the grasses Plant Physiol 135, 1776–1783

24 Fray RG & Grierson D (1993) Identification and genetic analysis of normal and mutant phytoene syn-thase genes of tomato by sequencing, complementation and co-suppression Plant Mol Biol 22, 589–602

25 Giorio G, Stigliani AL & D’Ambrosio C (2007) Agro-nomic performance and transcriptional analysis of carotenoid biosynthesis in fruits of transgenic HighCaro and control tomato lines under field conditions Trans-genic Res 16, 15–28

26 Giuliano G, Bartley GE & Scolnik PA (1993) Regula-tion of carotenoid biosynthesis during tomato develop-ment Plant Cell 5, 379–387

27 Zhang J (2003) Evolution by gene duplication: an update Trends Ecol Evol 18, 292–298

28 ABI PRISM 7700 Sequence Detection System (2001) Relative Quantitation of Gene Expression User Bulletin

#2 AB Applied Biosystems, Foster City, CA