Labellum transcriptome reveals alkene biosynthetic genes involved in orchid sexual deception and pollination-induced senescence

1 23

Functional & Integrative Genomics ISSN 1438-793X Funct Integr GenomicsDOI 10.1007/s10142-012-0288-x

Labellum transcriptome reveals alkenebiosynthetic genes involved in orchidsexual deception and pollination-inducedsenescence

Filipa Monteiro, Mónica Sebastiana,Andreia Figueiredo, Lisete Sousa, HelenaC. Cotrim & Maria Salomé Pais

1 23

Your article is protected by copyright and

all rights are held exclusively by Springer-

Verlag. This e-offprint is for personal use only

and shall not be self-archived in electronic

repositories. If you wish to self-archive your

work, please use the accepted author’s

version for posting to your own website or

your institution’s repository. You may further

deposit the accepted author’s version on a

funder’s repository at a funder’s request,

provided it is not made publicly available until

12 months after publication.

ORIGINAL PAPER

Labellum transcriptome reveals alkene biosynthetic genesinvolved in orchid sexual deception and pollination-inducedsenescence

Filipa Monteiro & Mónica Sebastiana &

Andreia Figueiredo & Lisete Sousa & Helena C. Cotrim &

Maria Salomé Pais

Received: 10 April 2012 /Revised: 17 May 2012 /Accepted: 28 May 2012# Springer-Verlag 2012

Abstract One of the most remarkable pollination strategy inorchids biology is pollination by sexual deception, in whichthe modified petal labellum lures pollinators bymimicking thechemical (e.g. sex pheromones), visual (e.g. colour and shape/size) and tactile (e.g. labellum trichomes) cues of the receptivefemale insect species. The present study aimed to characterizethe transcriptional changes occurring after pollination in thelabellum of a sexually deceptive orchid (Ophrys fusca Link) inorder to identify genes involved on signals responsible forpollinator attraction, the major goal of floral tissues. Novelinformation on alterations in the orchid petal labellum geneexpression occurring after pollination demonstrates a reduc-tion in the expression of alkene biosynthetic genes using O.fusca Link as the species under study. Petal labellum

transcriptional analysis revealed downregulation of transcriptsinvolved in both pigment machinery and scent compounds,acting as visual and olfactory cues, respectively, important insexual mimicry. Regulation of petal labellum senescence wasrevealed by transcripts related to macromolecules breakdown,protein synthesis and remobilization of nutrients.

Keywords Ophrys fusca . Labellum . Sexual deception .

Transcriptome . Alkene biosynthetic gene . Senescence

AbbreviationsaRNA Antisense RNADAP Days after pollinationDMSO Dimethyl sulfoxideMIPS Munich Information Center for Protein SequencesNCBI National Centre for Biotechnology InformationqPCR Quantitative real-time PCRROS Reactive oxygen speciesSAD Stearoyl acyl carrier protein (ACP) desaturaseSDS Sodium dodecyl sulphateSSC Sodium chloride–sodium citrate buffer

Introduction

Sexually deceptive orchids of the genus Ophrys are a primeexample of pollination by mimicry in plants (Schiestl 2005).The labellum, a modified petal often forming a landing plat-form for insects, displays visual, tactile and olfactory signalsfor pollinator’s attraction (Fay and Chase 2009; Schiestl 2005;Schiestl et al. 1997). In sexual deception, the orchids mimicthe sex pheromone of receptive female insects, thereby de-ceivingmale pollinators that attempt to mate with the labellumin a process known as pseudocopulation, whereby pollen is

Electronic supplementary material The online version of this article(doi:10.1007/s10142-012-0288-x) contains supplementary material,which is available to authorized users.

F. Monteiro (*) :M. Sebastiana :A. Figueiredo :H. C. Cotrim :M. S. PaisPlant Systems Biology Lab, Center for Biodiversity,Functional & Integrative Genomics (BioFIG), Faculty of Sciences,University of Lisbon,Ed. C2 Floor 4/2.4.48, Campo Grande 1749-016 Lisbon, Portugale-mail: [email protected]

L. SousaDepartment of Statistics and Operations Research, FCUL,CEAUL,Lisbon, Portugal

Present Address:H. C. CotrimNational Museum of Natural History,Botanic Garden and Centre for Environmental Biology (CBA),Faculty of Sciences, University of Lisbon,R. Escola Politécnica 58, ,1250-102 Lisbon, Portugal

Funct Integr GenomicsDOI 10.1007/s10142-012-0288-x

Author's personal copy

transferred (Kullenberg 1961; Schiestl et al. 2000).Additionally, labellum trichomes pattern mimic insect hairs,controlling the correct pollinator alignment with the flowerduring the copulation attempt (Ǻgren et al. 1984), and bothlabellum colour and shape imitate the insect female bodyparts. Pollination by sexual deceit is highly specific, witheach Ophrys species being pollinated by only one or afew related species of insects (Kullenberg 1961; Paulusand Gack 1981; Schiestl and Ayasse 2002; Schiestl et al.1999). Several studies on Ophrys floral signals led to therecognition of odour as the main cue for pollinator spec-ificity (Ayasse et al. 2011; Schiestl 2005; Schiestl et al.1999; Vereecken et al. 2010). This pollinator specificitymechanism is responsible for the reproductive isolation(floral isolation) among sympatric Ophrys species by ef-fectively preventing gene flow (Schiestl and Ayasse 2002;Vereecken et al. 2010; Xu et al. 2011). The blend ofsubstances that mimic the pollinator female sex pheromoneis composed of cuticular hydrocarbons, alkanes andalkenes (Schiestl et al. 1999). It was shown that twoclosely related Ophrys species, pollinated by different in-sect species, differ in the proportion of cuticular alkenes(Mant et al. 2005), leading to major odour differences andattraction of different pollinators, suggesting that alkene’sbiosynthetic genes may be candidate barrier genes(Schlüter et al. 2011). Recently, the expression pattern ofan Ophrys desaturase encoding gene, stearoyl acyl carrierprotein desaturase 2 (SAD2), was found to be related tothe production of a species-specific alkene bouquet, con-tributing to the differential pollinator attraction and repro-ductive isolation among Ophrys species (Schlüter et al.2011). Ophrys fusca Link (Schrader 1800), used in thepresent study, is a species native to the Mediterranean.

Flower function relies essentially on attracting pollinators toachieve successful pollination. Upon pollination, removal ofpollinated flowers of the plant or inflorescence is crucial so thatit does not compete for pollinators with the remaining flowersof the same inflorescence. Besides, the flower can be a sub-stantial sink on the plant’s resources and as such is energeti-cally expensive to maintain after its role has been fulfilled.Pollination sets off a cascade of developmental events thatinclude perianth senescence, changes in pigmentation, ovuledifferentiation and ovary maturation (Attri et al. 2007; O'Neill1997; O'Neill and Nadeau 1997). Petal senescence is a dynam-ic developmental process which involves highly coordinatedchanges in gene expression and requires active gene transcrip-tion and protein translation (Chapin and Jones 2007, 2009,Jones 2008). Also, there is significant evidence that nutrientremobilization is the central role of petal senescence followingpollination, as many senescence-enhanced genes in petalsencode for catabolic enzymes involved in the breakdownof macromolecules and cell organelles (van Doorn andWoltering 2008).

Microarray technology has been used in our study todetermine the expression pattern of thousands of genessimultaneously, thus producing a global overview of themolecular changes occurring at the transcriptional level, ina tissue or a specific developmental stage (Shena 2002). Thepresent study aimed to characterize the transcriptionalchanges occurring after pollination in the labellum of asexually deceptive orchid (O. fusca Link) in order to iden-tify genes involved on signals responsible for pollinatorattraction, the major goal of floral tissues. Further, scent-related genes (SADs) are discussed as barriers to cross-pollination. The data also identified several genes involvedin the timely initiation and progression of the floral senes-cence program.

Materials and methods

Plant material

O. fusca subsp. fusca (Pedersen and Faurholdt 2007) is awidely distributed species in Portugal and in the MediterraneanBasin. Labella were collected from a natural population at Serrada Arrábida Natural Park, Setúbal (N 38°30.43′11″, W 8°55.39′34″). One hundred plants were selected at anthesis stage. Plantswere covered with a disposable net (mesh size, 1×1 mm) toprevent natural pollination (Fig. 1a), considering that the aver-age body length of O. fusca s. l. pollinator is 11.9 mm (Paulus1997). During the flowering period (March 2007), 50 labellafrom 50 different individuals were cross-pollinated with a ster-ile plastic stick, and the remaining 50 were left unpollinated.Previous studies in Ophrys sphegodes labellum revealed odourand colour changes 2 and 4 days after pollination (DAP)(Schiestl and Ayasse 2001; Schiestl et al. 1997). Labella fromunpollinated and pollinated flowers were collected by cuttingthe lips at the stigmatic surface boundary at 2DAP and 4DAPduring late-morning. Five biological replicates were col-lected at each time point, each biological replicate beingrepresented by five labella from five different plants(Fig. 1b). Plant material was frozen and placed in driedice during fieldwork until storage at −80 °C. Pollinationsuccess was evaluated by inspecting capsule formation2 months later (data not shown).

RNA extraction and construction of cDNA libraries

Total RNA from O. fusca labella was extracted followingthe hot borate method (Wan and Wilkins 1994) with minormodifications as described in Sebastiana et al. (2009).Messenger RNA (mRNA) was purified using FastTrack®

MAG mRNA Isolation Kit (Invitrogen, Paisley, UK)according to the manufacturer’s instructions. Two cDNAlibraries, from labella of pollinated and unpollinated

Funct Integr Genomics

Author's personal copy

flowers, were constructed using CloneMinerTM cDNA li-brary construction kit (Invitrogen, Paisley, UK).

Construction of cDNA microarray

Randomly selected cDNA library clones were PCR-amplifiedas described previously (Figueiredo et al. 2008). A total of 3,384cDNA clones (1,692 clones from each cDNA library) wereselected by the presence of a unique PCR band higher than500 bp in agarose gels. Amplified cDNA clones were purifiedon Multiscreen®PCR 96 plates (Millipore, Bedford, MA, USA),transferred to V-bottom printing plates (Greiner-Corning, NY,USA), resuspended in printing buffer (50 % DMSO, 0.4×SSC)and printed in duplicate onto UltraGAPSTM-coated slides(Corning, USA) using VersArray ChipWriterTM CompactArrayer ver.3.1 (Bio-Rad Lab., Hercules, CA, USA).Technical details of spotting are provided according toMinimum Information About a Microarray Experiment(MIAME) guidelines (Online Resource 1 in “Electronic supple-mentary material”). After printing, cross-linking was performedwith heat and UV according to Vodkin Laboratory protocol[http://soybeangenomics.cropsci.uiuc.edu]. Microarray qualitywas assessed by Gel Star® Nucleic Acid Stain (FMC,Rockland,ME, USA) following themanufacturer’s instructions.

RNA amplification, labelling and hybridization

RNA amplification was achieved for all biological repli-cates. One microgram of total RNA from each biologicalreplicate was used to synthesize antisense RNA (aRNA)with MessageAmpTM II aRNA Amplification Kit(Ambion, Applied Biosystems) according to the manufac-turer’s instructions. Two micrograms of aRNA from eachbiological replicate was reverse-transcribed and labelledwith RPN 5660 CyScribeTM cDNA Post Labeling Kit(Amersham, GE Healthcare). Afterwards, complementaryDNA of labella from unpollinated flowers were labelled

with Cy3-dUTP (control) and labella of pollinated flowerswere labelled with Cy5-dUTP (test). Labelled cDNAs werepurified with CyScribeTM GFXTM Purification Kit(Amersham, GE Healthcare) according to the manufac-turer’s instructions. Hybridization and slide pre- and post-hybridization washes were done according to Sebastiana etal. (2009). For each time point, seven hybridizations werecarried out, matching five biological replicates and twotechnical replicates. A total of 14 hybridizations wereperformed.

Signal detection and microarray analysis

Image acquisition, spot and background quantification anddata normalization were done according to Figueiredo et al.(2008). Differentially expressed genes were statistically iden-tified with Rank Product method (Breitling et al. 2004) ran inBioconductor using the RankProd package ver. 2.18.0 with1,000 balanced permutations (http://www.bioconductor.org/packages/release/bioc/html/RankProd.html). Genes were con-sidered as differentially expressed when presenting a falsediscovery rate ≤5 % and ≥1.3 fold change.

Sequence annotation

Two hundred seventy-seven differentially expressed se-quenced tags (ESTs) were sequenced using BigDye termi-nator v3.1 cycle Sequencing Kit (Applied Biosystems,USA) in ABI PRISM 310 Genetic Analyser (AppliedBiosystems, USA) according to the manufacturer’s instruc-tions. Assembly of the individual ESTs was performed onSeqManII 5.0 software (DNAStar) (90 % similarity over 40-nt length). Contigs were named according to the clone withlowest E-value number. Sequences were annotated (OnlineResource 2 in “Electronic supplementary material”) by que-ry against public protein databases: NCBI non-redundantBasic Local Alignment Search Tool (BLAST)×(Altschul et

A B

Fig. 1 Plant material a O. fusca covered by a net disposable in the field; b experimental design

Funct Integr Genomics

Author's personal copy

al. 1997) and UniRef (UniProt Reference Clusters) 50(Suzek et al. 2007), setting an E-value of 10−5 as threshold(June 2011). Gene ontology (GO) (Ashburner et al. 2000)classification was accomplished by accessing GO annota-tion files resulting from the obtained UniRef50 accessionhomologies on Unref50 database. Annotation results wereintegrated with GO ‘Biological Process’ terms and MIPSFunctional Catalogue (FunCat) Database classification(Ruepp et al. 2004), supported by evidence from the litera-ture. Sequences were assigned into functional categories.

Quantitative real-time PCR: sample preparation, referencegenes and data analysis

Quantitative real-time PCR (qPCR) was performed with thesame RNA samples used in cDNA microarray hybridiza-tions, following Minimum Information for Publication ofQuantitative Real-Time PCR Experiments (MIQE) guide-lines (Bustin et al. 2009). Five biological replicates with twotechnical replicates were achieved. Relative quantificationwas performed following the standard curve method(Applied Biosystems 2008).

aRNA from labella of pollinated and unpollinated flowerswas DNase-treated with Turbo DNase-free Kit (Ambion,Applied Biosystems, USA) following the manufacturer’sinstructions. First-strand cDNA was synthesized from 2 μgaRNA with an N6 randomized primer, using RevertAid M-MuLV H− Reverse Transcriptase (Fermentas), as per suppli-er’s protocol. Diluted cDNA was amplified in a 25-μl qPCRreaction using MaximaTM SYBR Green qPCR Master Mix(2x) kit (Fermentas, Ontario, Canada) and 200 mM of eachprimer on a StepOneTM Real Time PCR system (AppliedBiosystems, Foster City, CA, USA). Primer sequences,designed using Primer Express ver. 1.0 software (AppliedBiosystems, USA), and product sizes for surveyed genes arelisted at Online Resource 3 in “Electronic supplementarymaterial”. Each set of reactions included a positive control(cDNA clone) and no template control. The amplificationprotocol was set at 95 °C for 10 min followed by 45 cyclesat 95 °C for 15 s, annealing temperature for 30 s, and plateread. Non-specific PCR products were analysed by dissocia-tion curves (Online Resource 4 in “Electronic supplementarymaterial”). Amplification efficiency was calculated from theslope of each standard curve (Online Resource 5 in“Electronic supplementary material”). Three genes were se-lected as possible reference genes from the RankProduct test:40 S ribosomal protein S10-like; oligopeptidase, predicted and4-α-glucanotransferase. NormFinder software (Andersen et al.2004) was used to select the most stable candidate genes andstability (M) values ≤0.5 were set as threshold (OnlineResource 6 in “Electronic supplementary material”).Afterwards, a normalization factor (NF) was calculated bythe geometric mean of the expression levels of the best

performing reference genes (Online Resource 7 in“Electronic supplementary material”) (Vandesompele et al.2002). Target gene expression (fold change) was calculatedby dividing experimental (pollinated replicates) with control(unpollinated replicates) datasets at time points analysed.

Results

Gene expression analysis

Differential gene expression profiling in O. fusca at 2 and 4DAP was assessed through cDNA microarray comparisonof labella from pollinated against unpollinated flowers.Microarray data have been deposited in NCBI’s GeneExpression Omnibus (GEO) database (Edgar et al. 2002)under accession number GSE28273. Statistical analysisrevealed 277 differentially expressed transcripts at both 2and 4 DAP (8.2 % of total cDNA clones printed).Considering the two time points under study, differentiallyexpressed genes increased from 106 in 2 DAP to 171 at 4DAP. Differentially expressed genes were generally upregu-lated (166 transcripts, 59.9 %), with 54 transcripts (19.5 %)being found at 2 DAP and 112 transcripts (40.4 %) at 4DAP.From the total downregulated (40.1 %) transcripts (111 tran-scripts), 52 (18.8 %) were found at 2 DAP and 59 (21.3 %)at 4 DAP. Transcripts assemblage revealed 140 unigenes(50.5 %) corresponding to 114 singletons and 26 contigs(Online Resource 2 in “Electronic supplementary material”).Twenty-four unigenes were shared between 2 and 4 DAP, 11being upregulated and six downregulated. Unigenes weregrouped into 13 functional categories. The most prevalentfunctional categories include no homology (24 %), repre-senting not yet identified genes, and unclassified (19 %)categories, followed by metabolism and cell rescue, defenseand virulence (Fig. 2). Although no virus-infected symp-toms were found in the analysed samples, transcripts codingfor non-plant and viral proteins were detected (OnlineResource 2 in “Electronic supplementary material”).Transcripts coding for viral proteins have been previouslydetected on Phalaenopsis sp. flower buds (Hsiao et al. 2011;Tsai et al. 2006).

Validating microarray expression data by qPCR

To validate microarray gene expression profiles, six differen-tially expressed genes encoding for pathogenesis-related pro-tein 10c (HO849917), predicted protein (HO849934), non-lysosomal glucosylceramidase (HO850028), stearoyl-acyl-carrier protein desaturase (HO849909), metallothionein-likeprotein type 3 (HO850065) and stearoyl-ACP desaturase ho-mologue 2 (HO849908) were quantified by qPCR. Referencegenes encoding 40 ribosomal protein S10-like and

Funct Integr Genomics

Author's personal copy

oligopeptidase displayed the best combination gene stabilityvalue (Online Resource 6 in “Electronic supplementary mate-rial”). The general trend of qPCR expression profiles wassimilar to that obtained by cDNA microarray analysis (Fig. 3).

Discussion

A cDNA microarray approach was conducted to address la-bellum transcriptional changes at 2 and 4 DAP. Comparativetranscriptional analysis of O. fusca labellum from pollinatedand unpollinated flowers revealed major gene expression in-cluding: activation of stress and defense responses,

macromolecules breakdown (proteolysis), nutrient remobiliza-tion (phosphate, carbon and nitrogen), deactivation of photo-synthetic machinery and of secondary metabolism pathwaysinvolved in pigmentation and scent production and en-hancement of de novo protein synthesis.

Transcriptional regulation of labellum senescence

Pollination-regulated events collectively prepare the flower forfertilization and embryogenesis while bringing about the lossof floral organs that have completed their function in pollendispersal and reception (O'Neill 1997). Such events includechanges in flower pigmentation, senescence and abscission of

Protein synthesis

Cellular Comm/Signal transd

Cellular transport/facilities and routes

Transc. Factors and Regul.Transc.

Biog cellular components

Protein Fate

Non-plant, viral and plasmid proteins

Energy

Cell Fate

Cell Resc, Def and Virul

Metabolism

Unclassified

No homology 24%

19%

15%

10%

8 %

6 %

5 %

4 %

3 %

2 %

2 %

1 %

1 %

Fig. 2 Functional classificationof differentially expressedunigenes, considering both 2and 4 DAP time points.Classification followed the GOand MIPS FunCat annotationssupported by evidences fromthe literature. Percentage oftranscripts classified in each ofthe functional category isindicated

-6

-4

-2

0

2

4

qPCR Microarray

PR10 GlucosylCer PredictProt SAD2 Metallo III

2 D

AP

4 D

AP

2 D

AP

4 D

AP

2 D

AP

4 D

AP

4 D

AP

2 D

AP

4 D

AP

SAD1

Fig. 3 Evaluation of cDNA microarray and qPCR analysis for differen-tially expressed genes selected from microarray analysis. Transcripts cho-sen for validation encode for: pathogenesis- related protein 10c (PR10),glucosylceramidase (glucosylCer), predicted protein (PredictProt), stearoylACP-desaturase (SAD2), metallothionein type 3 (metallo III) and stearoylACP-desaturase (Ofup2825–SAD1). Median and mean absolute deviation

of five biological replicates, for both qPCR and microarrays, are presented.Two technical replicates were performed in all qPCR analysis. 40 S ribo-somal protein and oligopeptidase sequences were used as reference genesfor normalizing quantitative values. DAP days after pollination; FC FoldChange

Funct Integr Genomics

Author's personal copy

floral organs (Attri et al. 2007; O'Neill and Nadeau 1997).Overall, during flower perianth senescence, macromoleculesare degraded and organelles are dismantled, allowing the re-mobilization of nutrients from petals to other still developingtissues (Bai et al. 2010; Chapin and Jones 2007). This cellularmechanism is characterized by the expression of senescence-associated genes, encoding proteins such as cysteine proteases,RNases, ubiquitins, metallothioneins, glutathione S-transferaseand phosphatases (reviewed in Buchanan-Wollaston et al.2005; van Doorn and Woltering 2008). In O. fusca, transcrip-tional alterations in genes related to the senescence processwere detected as soon as 2 DAP, long before the detection ofany visual signs of labellum decline, a phenomenon onlydetected 5–6 days after pollination (data not shown). Thetranscriptional profile of several genes identified in our micro-array is consistent with a senescence process occurring afterpollination, namely, suggested by genes involved in stress anddefense responses, macromolecules breakdown, nutrient remo-bilization and de novo protein synthesis. This is in accordancewith numerous studies on pollination (Lan et al. 2004, 2005;Yu and Setter 2003) that report the induction of a senescenceprogram in floral tissues after pollination. Notably, severaltranscripts, representing 10 % of the differentially expressedunigenes (Fig. 2), associated to stress and defense responseswere activated after pollination, including pathogenesis-related10c protein, metallothioneins, antimicrobial snaking protein orglutathione S-transferase. This is in accordance with previousstudies that report the activation of stress and defense-relatedgenes accompanying petal senescence (Hoeberichts et al.2007; Price et al. 2008; vanDoorn et al. 2003), their expressionbeing involved on tissue protection against pathogen attackand reactive oxygen species (ROS) accumulation effects dur-ing the senescence process (Lan et al. 2004, 2005). The in-crease in the expression of genes encoding metallothioneinsduring petal senescence has been related to metal transport anddetoxification of ROS (Breeze et al. 2004; Hunter et al. 2002).Since metallothioneins prevent toxic levels of intracellular freecopper by binding to copper ions (Balamurugan and Schaffner2006; Hall 2002), which are released during protein degrada-tion, upregulation of several metallothionein transcripts at 4DAP may indicate ROS accumulation in O. fusca labellum.High representation of metallothioneins in the transcriptome ofsenescent petals has been reported (Breeze et al. 2004; Hunteret al. 2002; Wagstaff et al. 2010). Increased transcript abun-dance of glutathione S-transferases in O. fusca labellum afterpollination supports their role in protection against oxidativestress, which is a common event on petal senescence(Hoeberichts et al. 2007; Itzhaki et al. 1994; Rogers 2012).

Autophagy plays an important role in petal senescenceand it has been implicated in cell dismantling and remobi-lization of essential nutrients from senescing petals/sepals todeveloping floral parts (Hoeberichts et al. 2007; Shahri andTahir 2011a; van Doorn and Woltering 2008). In the

labellum of pollinated O. fusca flowers, activation of differ-ent sets of transcripts predicted to be involved on proteindegradation and catabolism of allantoate and fatty acidssuggests that macromolecule breakdown and nutrient remo-bilization is occurring. Activation of genes involved inprotein degradation at 2 DAP, and to larger extent at 4DAP, such as cysteine proteases and polygalacturonase,suggests an increase in proteolytic activity, possibly allow-ing nitrogen (N) remobilization (Jones et al. 2005). Genespredicted to be involved on remobilization of nutrients wereupregulated after pollination, including allantoate amidohy-drolase, which allows the plant to recover stored N, carbon(C) and energy (Werner et al. 2010) and 3-ketoacyl-CoAthiolase, involved in the final step of fatty acids β-oxidation(Castillo and Leon 2008). During petal senescence, degra-dation of free fatty acids mainly occurs by β-oxidation(Hoeberichts et al. 2007; Pistelli et al. 1991), yielding ace-tyl–CoA which is further used in the glyoxylate cycle,producing organic acids which can be converted to sugars(gluconeogenesis) (Chen et al. 2000; Cornah et al. 2004) orbe used for the synthesis of amino acids for protein synthe-sis (van Doorn and Woltering 2008). An increased transcriptabundance of several other genes involved in amino acidssynthesis has been identified in senescence petals (for a revi-sion, see van Doorn and Woltering (2008)). In this context, inO. fusca labellum, several transcripts predicted to be involvedin amino acids and protein synthesis were found to be upregu-lated at 4 DAP, such as phosphoglycerate dehydrogenase in-volved on serine synthesis (Ho et al. 1999), ribosomal subunitL17 protein (Meng et al. 2010) and translation initiation factorprotein (Fletcher et al. 1999). The treatment of flowers withcompounds that inhibit protein synthesis has been found todelay petal senescence, revealing the requirement of a newgene transcription and de novo protein synthesis to generatesuicidal proteins, such as proteases, which may be newlytranslated during senescence and/or activated by the newlytranslated proteins (Islam et al. 2011; Jones et al. 1994; Shahriand Tahir 2010, 2011b) . Upregulation, at 4 DAP, of a phos-phoglycerate dehydrogenase transcript (EC 1.1.1.95), involvedin the phosphorylated pathway of serine synthesis (Ho et al.1999), was observed. Serine (Ser) is an important precursor oftryptophan and cysteine amino acids (Walton and Woolhouse1986) and is part of the proteolytic serine proteases that areactively involved on senescence (Azeez et al. 2007; Pak andvan Doorn 2005). As a result, the increased serine synthesis inO. fusca labellum after pollination may probably assist the newprotein synthesis, a characteristic of the senescence program.

In pollination-induced petal senescence, the increasedexpression of RNases and DNases has been related tonucleic acid degradation (Panavas et al. 2000; Xu andHanson 2000). Two RNase transcripts were identified asupregulated at 4 DAP, suggesting RNA degradation afterpollination. These two transcripts are homologs to Prunus

Funct Integr Genomics

Author's personal copy

dulcis RNase PD2 transcripts which were associated withphosphate (Pi) remobilization during senescence (Ma andOliveira 2000; Price et al. 2008), thus suggesting that in O.fusca labellum these transcripts may be also involved on Pimobilization through nucleic acid degradation.

Lipid-related transcripts were downregulated in the label-lum after pollination, namely, those involved on ceramidebiosynthetic pathway. The observations that ceramide, asimple sphingolipid, may be involved in regulating plantPCD have led to efforts to understand how ceramides maybe metabolized through the action of ceramidases (CDases)(Pata et al. 2010). Ceramide is a well-established inducer ofapoptosis in animal cells (Hannun and Obeid 2008), and inplants, evidences of a similar role have also been reported(Liang et al. 2003; Shi et al. 2007). In O. fusca labellum,downregulation of a ceramidase was observed, suggestingceramide accumulation after pollination. As a result, aceramide-induced PCD could be occurring after pollination.

In addition to the green leaves, plants can potentially usereproductive structures to perform photosynthetic CO2, beingthe carbon supplied from floral photosynthesis a significantcontribution to total carbon (Aschan and Pfanz 2003), as amechanism to compensate for the costs of reproduction(Aschan and Pfanz 2006). Thus, it was not surprising to detectdownregulation of photosynthetic-related genes in O. fuscalabellum after pollination at 4 DAP, including light harvestingcomplex I, the photosystem I subunit G and RuBisCOactivase. The species under study, O. fusca, has a brownishlabellum with a green colour at the abaxial part. Despite theflower organ functions mainly as a sink, our results indicatethat the photosynthetic apparatus is operative and maycontribute to C supply (Müller et al. 2010), supporting thecosts of reproduction. Mitochondrion is the main organ ofenergy production, mostly in the form of ATP via oxidativephosphorylation and also a major site of intracellular ROSgeneration. ATP synthesized in the mitochondria is exchangedfor cytosolic ADP by adenine nucleotide translocator (ANT)to provide a continuous supply of ADP to the mitochondria(Haferkamp et al. 2011). Upregulation of a mitochondrialANT transcript at 4 DAP may be related to an increase onADP/ATP exchange that could ultimately cause ROS produc-tion (Laloi 1999). Despite the prime function of ANT inoxidative phosphorylation across the inner mitochondrialmembrane, ANT is a protein of the permeability transitionpore (PTP) complex (Crompton 1999). PTP activity has beenassociated to an apoptotic-like PCD in plants (Reape andMcCabe 2010), and ANT involvement in senescence cannotbe excluded.

Inmany species, pollination induces an ethylene burst whichcoordinates flower senescence (Jones 2008; van Doorn 2001),which is characteristic of the ethylene-sensitive flowers, asmany orchids are (O'Neill et al. 1993). Pollination-inducedethylene production and the roles of ethylene biosynthetic

pathway genes (1-aminocyclopropane-1-carboxylic acid(ACC) synthase (ACS) and ACC oxidase (ACO)) have beendemonstrated in pollinated (Bui and O'Neill 1998; Llop-Tous etal. 2000; O'Neill et al. 1993) and senescent (Henskens et al.1994; Woodson et al. 1992) flowers. O'Neill et al. (1993)verified that post-pollination events in orchids involved coor-dinated interorgan regulation of both ACS and ACO expression.On the labellum of Phalaenopsis flowers, an increased expres-sion of both ACS and ACO genes 24 h after pollination wasobserved, with a subsequent decrease on ACS and maintenanceof ACO 48 h after pollination (ACO, Nadeau et al. 1993; ACS,Bui and O'Neill 1998). As a result, in O. fusca labellum, it wasrather unexpected not to detect transcripts coding for ethylenebiosynthetic enzymes in pollinated flowers (O'Neill et al. 1993).

Attraction and deterrence: adaptation for achievingreproductive success

Flowers face a dynamic tension regarding the production ofspecific compounds that act as attractants to pollinators,while others act as deterrents to herbivores (Raguso 2008;Schiestl 2010; van der Meijden 1996). Ophrys labellummajor pigments are anthocyanins (Strack et al. 1989), whichare synthesized by the flavonoid biosynthetic pathway alongthe phenylpropanoid metabolism (Richard et al. 2000; Zhouet al. 2011). Downregulation of a chalcone synthase tran-script, encoding an enzyme that functions on anthocyaninsbiosynthesis (Richard et al. 2000), suggests that, asexpected, pigment biosynthetic machinery, important as vi-sual cue to pollinators, is no longer required after successfulpollination. Alkaloids include substances such as caffeineand morphine, and its toxicity and bitter taste are thought toplay a role in plant defense to herbivores and pathogens(Adler et al. 2001, Howe and Jander 2008). Downregulationof a 3′-N-debenzoyl-2′-deoxytaxol N-benzoyltransferasetranscript (EC. 2.3.1.150), involved in morphine synthesis(Grothe et al. 2001), suggests that alkaloids are important inO. fusca labellum as a barrier to florivores, its biosynthesisbeing inhibited after successful pollination. Since damagedtissues, such as petals, are less prone to pollinator visitation(Botto-Mahan et al. 2011), alkaloid synthesis would consti-tute a chemical defence to reduce florivory events untilpollination has occurred.

Scent-related genes with distinct functions

For pollinator attraction, Ophrys flowers chemically mimicthe pollinators’ female sex pheromones, being alkenes withdifferent chain lengths and double bond positions, the keycompounds for attracting the specific insect species.Recently, Schlüter et al. (2011), working with two co-flowering Ophrys species, O. sphegodes and O. exaltata,each pollinated by different insect species, identified a

Funct Integr Genomics

Author's personal copy

flower-specific gene encoding for a stearoyl-acyl carrierprotein desaturase (SAD2), involved in the biosynthesis of9- and 12-alkenes. SAD2 expression differences in the twoOphrys species are related to the pollinator specificity ob-served. Higher expression of SAD2 in O. sphegodes than inO. exaltata results in higher levels of 9- and 12-alkenesworking as attractants to the bee Andrena nigroaenea, thespecific pollinator of O. sphegodes. Conversely, the beeColletes cunicularius (O. exaltata pollinator) is attractedby 7-alkenes (of unknown source), whereas 9-alkenes re-duce this attraction. By contributing to the production ofspecies-specific alkene bouquets, SAD2 is suggested to be abarrier gene responsible for differential pollinator attractionand reproductive isolation among these species. In ourmicroarray, two different transcripts (Online Resource 2 in“Electronic supplementary material”) downregulated bypollination revealed homology with SADs proteins:Ofctg2559 coding for a SAD2 homolog and Ofup2825, withhomology to a SAD from Elaeis oleifera. Downregulationin O. fusca labellum after pollination of a SAD homolog(Ofup2825) can be related to a shutdown of fatty acidsynthesis after pollination as a consequence of senescence-related changes on cell membranes rather than a role onalkene’s biosynthesis. SAD is a stearoyl-acyl carrier proteindesaturase involved on oleic acid synthesis (Byfield andUpchurch 2007). This is further corroborated by the co-expression of a transcript homolog to a ω-6 fatty aciddesaturase involved in linoleic acid synthesis. ConcerningSAD2, its higher transcriptional levels in O. fusca unpol-linated labellum is suggestive of a role on 9- and 12-alkene biosynthesis. O. fusca, like O. sphegodes, is pol-linated by A. nigroaenea bees (Paulus and Gack 1990).Since 9- and 12-alkenes are key compounds for A.nigroaenea pollinator attraction, for both O. fusca(Schiestl and Ayasse 2002) and O. sphegodes (Schiestlet al. 2000), SAD2 is responsible for the biosynthesis ofcompounds involved on the specific attraction of theirpollinator (Schlüter et al. 2011). Thus, downregulation at4 DAP of a SAD2 may be related to a decrease onalkene production, which are important compounds asolfactory cues. Chemical analysis on O. sphegodes label-lum extracts 3 days after pollination revealed a slightdecrease on key odour compounds (Schiestl and Ayasse2001). Given the low volatility of alkenes, they mayremain on the labellum even when production is reduced(Schiestl and Ayasse 2001). Also, these compounds arepart of the plant epicuticular wax important for waterloss prevention (Hadley 1981; Jetter and Kunst 2008)and its immediate cessation could cause physiologicalconstraints (Schiestl and Ayasse 2001). Besides, mainte-nance of the inflorescence attractiveness even after suc-cessful pollination may increase the reproductive successof the whole inflorescence (Schiestl and Ayasse 2001),

which is particularly important in specialized pollinationsystems with low visitation rates as sexually deceptiveorchids (Schiestl and Schlüter 2009).

In Andrena-pollinated Ophrys, as species in O. fuscacomplex (Paulus 2001), flowers attract their pollinators bythe same co-occurring alkenes in different proportions(Schiestl 2005; Schiestl and Ayasse 2002; Schiestl et al.1999; Stökl et al. 2005, 2007). This bouquet-based speci-ficity of pollinator attraction is an important precondition forspeciation in the O. fusca-lutea species group (Stökl et al.2008, 2009). Thus, a slight change on the alkenes’ relativeamounts could be responsible for the attraction of a differentAndrena species. Chemical scent profiling, together withexpression analysis on SAD2 gene, could provide data ondifferences within O. fusca complex species. This couldprovide information on species-specific scent profile andscent genes, which are direct targets for selection inpollinator-mediated isolation.

Overall, transcriptional analysis in the sexually deceptiveorchid O. fusca allowed to access gene expression patternsoccurring in the labellum after pollination. Major findingspoint out to a repression of secondary metabolism pathwaysinvolved in labellum traits important for providing cues topollinators in sexual deception strategy (e.g. pigmentationand scent production) and activation of several molecularpathways associated to the pollination-induced senescenceof the labellum (i.e. macromolecules breakdown, non-specific stress pathways, nutrient remobilization and newprotein synthesis). The data obtained with this transcription-al approach widely contribute to the understanding of themolecular events following pollination in an orchid with apeculiar pollination strategy.

Acknowledgments We thank F.P. Schiestl for critical reading of themanuscript and helpful suggestions, V. Pereira, S. Serrazina, A.S. Róisand F. Pessoa for fieldwork assistance, A.M. Fortes for helping withGEO data submission and S. Ferreira and M. Romeiras for discussions.This work was supported by the Portuguese Foundation for Scienceand Technology with the fellowships SFRH/BD/30152/2006, FCT/OE,FCT/Ciência 2007 and BIOFIG PEst-OE/BIA/UI4046/2011.

References

Adler L, Karban R, Strauss S (2001) Direct and indirect effects ofalkaloids on plant fitness via herbivory and pollination. Ecol82:2032–2044

Ǻgren L, Kullenberg B, Sensenbaught T (1984) Congruences in pilositybetween three species of Ophrys (Orchidaceae) in their hymenop-teran pollinators. Nova Acta Reg Soc Sci Ups Ser V 3(3):15–25

Altschul S, Madden T, Schäffer A, Zhang J, Zhang Z, Miller W,Lipman D (1997) Gapped BLAST and PSI-BLAST: a new gen-eration of protein database search programs. Nucleic Acids Res25:3389–3402

Andersen C, Jensen J, Orntoft T (2004) Normalization of real-timequantitative reverse transcription-PCR data: a model-based

Funct Integr Genomics

Author's personal copy

variance estimation approach to identify genes suited for normal-ization, applied to bladder and colon cancer data sets. Cancer Res64:5245–5250

Applied Biosystems (2008) Guide to performing relative quantitationof gene expression using real-time quantitative PCR. Available athttp://www3.appliedbiosystems.com/cms/groups/mcb_support/documents/generaldocuments/cms_042380.pdf

Aschan G, Pfanz H (2003) Non-foliar photosynthesis—a strategy ofadditional carbon acquisition. Flora 198:81–97

Aschan G, Pfanz H (2006) Why snowdrop (Galanthus nivalis L.)petals have green marks? Flora 201:623–632

Ashburner M, Ball C, Blake J, Botstein D, Butler H, Cherry J, Davis A,Dolinski K, Dwight S, Eppig J (2000) Gene Ontology: tool for theunification of biology. The Gene Ontology consortium. Nat Genet25(1):25–29

Attri L, Nayyar H, Bhanwra R, Vij S (2007) Post-pollination biochem-ical changes in the floral organs of Rhynchostylis retusa (L.) Bl.and Aerides multiflora Roxb. (Orchidaceae). J Plant Biol 50:548–556

Ayasse M, Stokl J, Francke W (2011) Chemical ecology andpollinator-driven speciation in sexually deceptive orchids.Phytochem 72:1667–1677

Azeez A, Sane A, Bhatnagar D, Nath P (2007) Enhanced expression ofserine proteases during floral senescence in Gladiolus. Phytochem68:1352–1357

Bai S, Willard B, Chapin L, Kinter M, Francis D, Stead A, Jones M(2010) Proteomic analysis of pollination-induced corolla senes-cence in petunia. J Exp Bot 61:1089–1109

Balamurugan K, Schaffner W (2006) Copper homeostasis in eukar-yotes: teetering on a tightrope. Biochim Biophys Acta, Mol CellRes 1763:737–746

Botto-Mahan C, Ramírez PA, Ossa CG, Medel R, Ojeda-Camacho M,González AV (2011) Floral herbivory affects female reproductivesuccess and pollinator visitation in the perennial herbAlstroemeria ligtu (Astroemeriaceae). Int J Plant Sci 172(9):1130–1136

Breeze E, Wagstaff C, Harrison E, Bramke I, Rogers H, Stead A,Thomas B, Buchanan-Wollaston V (2004) Gene expression pat-terns to define stages of post-harvest senescence in Alstroemeriapetals. Plant Biotechnol J 2:155–168

Breitling R, Armengaud P, Amtmann A, Herzyk P (2004) Rank prod-ucts: a simple, yet powerful, new method to detect differentiallyregulated genes in replicated microarray experiments. FEBS Lett573:83–92

Buchanan-Wollaston V, Page T, Harrison E, Breeze E, Lim P, Nam H,Lin J, Wu S, Swidzinski J, Ishizaki K, Leaver C (2005)Comparative transcriptome analysis reveals significant differen-ces in gene expression and signalling pathways between develop-mental and dark/starvation-induced senescence in Arabidopsis.Plant J 42:567–585

Bui A, O'Neill S (1998) Three 1-aminocyclopropane-1-carboxylatesynthase genes regulated by primary and secondary pollinationsignals in orchid flowers. Plant Physiol 116:419–428

Bustin S, Benes V, Garson J, Hellemans J, Huggett J, Kubista M,Mueller R, Nolan T, Pfaffl M, Shipley G, Vandesompele J,Wittwer C (2009) The MIQE guidelines: minimum informationfor publication of quantitative real-time PCR experiments. ClinChem 55:611–622

Byfield G, Upchurch R (2007) Effect of temperature on delta-9stearoyl-ACP and microsomal omega-6 desaturase gene expres-sion and fatty acid content in developing soybean seeds. Crop Sci47:1698–1704

Castillo M, Leon J (2008) Expression of the beta-oxidation gene 3-ketoacyl-CoA thiolase 2(KAT2) is required for the timely onset ofnatural and dark-induced leaf senescence in Arabidopsis. J ExpBot 59:2171–2179

Chapin L, Jones M (2007) Nutrient remobilization during pollinationinduced corolla senescence in Petunia. Acta Horticult 755:181–190

Chapin LJ, Jones ML (2009) Ethylene regulates phosphorus remobili-zation and expression of a phosphate transporter (PhPT1) duringpetal senescence in petunia. J Exp Bot 60:2179–2190

Chen Z, Walker R, Acheson R, Tecsi L, Wingler A, Lea P, Leegood R(2000) Are isocitrate lyase and phosphoenolpyruvate carboxyki-nase involved in gluconeogenesis during senescence of barleyleaves and cucumber cotyledons? Plant Cell Physiol 41:960–967

Cornah J, Germain V, Ward J, Beale M, Smith S (2004) Lipid utiliza-tion, gluconeogenesis, and seedling growth in Arabidopsismutants lacking the glyoxylate cycle enzyme malate synthase. JBiol Chem 279:42916–42923

Crompton M (1999) The mitochondrial permeability transition poreand its role in cell death. Biochem J 341:233–249

Edgar R, Domrachev M, Lash A (2002) Gene Expression Omnibus:NCBI gene expression and hybridization array data repository.Nucleic Acids Res 30:207–210

Fay M, Chase M (2009) Orchid biology: from Linnaeus via Darwin tothe 21st century. Ann Bot 104:359–364

Figueiredo A, Fortes A, Ferreira S, Sebastiana M, Choi Y, Sousa L,Acioli-Santos B, Pessoa F, Verpoorte R, Pais M (2008)Transcriptional and metabolic profiling of grape (Vitis viniferaL.) leaves unravel possible innate resistance against pathogenicfungi. J Exp Bot 59:3371–3381

Fletcher C, Pestova T, Hellen C, Wagner G (1999) Structure andinteractions of the translation initiation factor eIF1. EMBO J 18(9):2631–2637

Grothe T, Lenz R, Kutchan T (2001) Molecular characterization of thesalutaridinol 7-O-acetyltransferase involved in morphine biosyn-thesis in opium poppy Papaver somniferum. J Biol Chem276:30717–30723

Hadley N (1981) Cuticular lipids of terrestrial plants and arthropods—a comparision of their structure, composition, and waterproofingfunction. Biol Rev Camb Philos Soc 56:23–47

Haferkamp I, Fernie A, Neuhaus H (2011) Adenine nucleotide trans-port in plants: much more than a mitochondrial issue. TrendsPlant Sci 16:507–515

Hall J (2002) Cellular mechanism for heavy metal detoxification andtolerance. J Exp Bot 53(366):1–11

Hannun Y, Obeid L (2008) Principles of bioactive lipid signalling:lessons from sphingolipids. Nat Rev Mol Cell Biol 9:139–150

Henskens J, Rouwendal G, Have A, Woltering E (1994) Molecularcloning of two different ACC synthase PCR fragments in carna-tion flowers and organ-specific expression of the correspondinggenes. Plant Mol Biol 26(1):453–458

Ho C, Noji M, Saito M, Saito K (1999) Regulation of serine biosynthesisin Arabidopsis—crucial role of plastidic 3-phosphoglycerate dehy-drogenase in non-photosynthetic tissues. J Biol Chem 274:397–402

Hoeberichts F, van Doorn W, Vorst O, Hall R, van Wordragen M(2007) Sucrose prevents up-regulation of senescence-associatedgenes in carnation petals. J Exp Bot 58:2873–2885

Howe GA, Jander G (2008) Plant immunity to insect herbivores. AnnuRev Plant Biol 59:41–66

Hsiao Y, Chen Y, Huang S, Pan Z, Fu C, Chen W, Tsai W, Chen H(2011) Gene discovery using next-generation pyrosequencing todevelop ESTs for Phalaenopsis orchids. BMC Genomics 12:360

Hunter D, Steele B, Reid M (2002) Identification of genes associatedwith perianth senescence in daffodil (Narcissus pseudonarcissusL. 'Dutch Master'). Plant Sci 163:13–21

Islam S, Tahir I, Shahri S, Bhat M (2011) Effect of cycloheximide onsenescence and postharvest performance in Hemerocallis fulva cv.Royal Crown. J Plant Sci 6:14–25

Itzhaki H, Maxson J, Woodson W (1994) An ethylene-responsiveenhancer element is involved in the senescence-related expression

Funct Integr Genomics

Author's personal copy

of the carnation glutathione-S-transferase (GST1) gene. PNASUSA 91:8925–8929

Jetter R, Kunst L (2008) Plant surface lipid biosynthetic pathways andtheir utility for metabolic engineering of waxes and hydrocarbonbiofuels. Plant J 54:670–683

Jones M (2008) Ethylene signalling is required for pollination-accelerated senescence in Petunia. Plant Sci 175:190–196

Jones R, Serek M, Kuo C-L, Reid M (1994) The effect of proteinsynthesis inhibition on petal senescence in cut bulb flowers. J AmSoc Hortic Sci 119(6):1243–1247

Jones M, Chaffin G, Eason J, Clark D (2005) Ethylene-sensitivityregulates proteolytic activity and cysteine protease gene expres-sion in petunia corollas. J Exp Bot 56:2733–2744

Kullenberg B (1961) Studies on Ophrys pollination. Zool Bidrag Upps34:1–340

Laloi M (1999) Plant mitochondrial carriers: an overview. Cell MolLife Sci 56:918–944

Lan L, Chen W, Lai Y, Suo J, Kong Z, Li C, Lu Y, Zhang Y, Zhao X,Zhang X, Zhang Y, Han B, Cheng J, Xue J (2004) Monitoringof gene expression profiles and isolation of candidate genesinvolved in pollination and fertilization in rice (Oryza sativaL.) with a 10 K cDNA microarray. Plant Mol Biol 54(4):471–487

Lan L, Li M, Lai Y, Xu W, Kong Z, Ying K, Han B, Xue Y(2005) Microarray analysis reveals similarities and variationsin genetic programs controlling pollination/fertilization andstress responses in rice (Oryza sativa L.). Plant Mol Biol59(1):151–164

Liang H, Yao N, Song L, Luo S, Lu H, Greenberg L (2003) Ceramidesmodulate programmed cell death in plants. Genes Dev 17:2636–2641

Llop-Tous I, Barry C, Grierson D (2000) Regulation of ethylenebiosynthesis in response to pollination in tomato flowers. PlantPhysiol 123:971–978

Ma R-C, Oliveira M (2000) The RNase PD2 gene of almond (Prunusdulcis) represents an evolutionary distinct class of S-like RNasegenes. Mol Gen Genet 263(6):925–933

Mant J, Peakall R, Schiestl F (2005) Does selection on floral odorpromote differentiation among populations and species of thesexually deceptive orchid genus Ophrys? Evol 59:1449–1463

Meng F, Zhang L, Kang M, Guo X, Xu B (2010) Molecular charac-terization, immunohistochemical localization and expression of aribosomal protein L17 gene from Apis cerana cerana. Arch InsectBiochem Physiol 75(2):121–138

Müller G, Drincovich M, Andreo C, Lara M (2010) Role of photosyn-thesis and analysis of key enzymes involved in primary metabo-lism throughout the lifespan of the tobacco flower. J Exp Bot61:3675–3688

Nadeau J, Zhang X, Nair H, O'Neill S (1993) Temporal and spatialregulation of 1-aminocyclopropane-1-carboxylate oxidase in thepollination-induced senescence of orchid flowers. Plant Physiol103(1):31–39

O'Neill S (1997) Pollination regulation of flower development. AnnuRev Plant Physiol Plant Mol Biol 48:547–574

O'Neill S, Nadeau J (1997) Postpollination flower development. HorticRev 19:1–58

O'Neill S, Nadeau J, Zhang X, Bui A, Halevy A (1993) Interorganregulation of ethylene biosynthetic genes by pollination. PlantCell 5:419–432

Pak C, van Doorn W (2005) Delay of Iris flower senescence byprotease inhibitors. New Phytol 165:473–480

Panavas T, LeVangie R, Mistler J, Reid P, Rubinstein B (2000)Activities of nucleases in senescing daylily petals. Plant PhysiolBiochem 38:837–843

Pata M, Hannun Y, Ng CK-Y (2010) Plant sphingolipids: decoding theenigma of the Sphinx. New Phytol 185:611–630

Paulus H (1997) Signale in der bestäuberanlockung: weibchenimita-tion als bestäubungsprinzip bei der mediterranen orchideengat-tung Ophrys. Verh Zool Bot Ges Österr 134:133–176

Paulus H (2001) Material zu einer Revision des Ophrys fusca s.str.Artenkreises. I. Ophrys nigroaenea-fusca, O. colletes-fusca, O.flavipes-fusca, O. funerea, O. forestieri oder was ist die typischeOphrys fusca Link 1799 (Orchidaceae)? J Eur Orch 33(1):121–177

Paulus H, Gack C (1981) Neue Beobachtungen zur Bestäubung vonOphrys (Orchidaceae) in Südspanien, mit besondererBerücksichtigung des Formenkreises Ophrys fusca agg. PlantSyst Evol 137:241–258

Paulus H, Gack C (1990) Pollination of Ophrys (Orchidaceae) inCyprus. Plant Syst Evol 169:177–207

Pedersen H, Faurholdt N (2007) Ophrys: the bee orchids of Europe.Royal Botanic Gardens, Kew, Hardcover, pp 40–91

Pistelli L, De Bellis L, Alpi A (1991) Peroxisomal enzyme activities inattached senescing leaves. Planta 184(1):151–153

Price A, Orellana D, Salleh F, Stevens R, Acock R, Buchanan-Wollaston V, Stead A, Rogers H (2008) A comparison of leafand petal senescence in wallflower reveals common and distinctpatterns of gene expression and physiology. Plant Physiol147:1898–1912

Raguso R (2008) Wake up and smell the roses: the ecology andevolution of floral scent. Annu Rev Ecol Evol Syst 39:549–569

Reape T, McCabe P (2010) Apoptotic-like regulation of programmedcell death in plants. Apoptosis 15(3):249–256

Richard S, Lapointe G, Rutledge R, Seguin A (2000) Induction ofchalcone synthase expression in white spruce by wounding andjasmonate. Plant Cell Physiol 41(8):982–987

Rogers H (2012) Is there an important role for reactive oxygen speciesand redox regulation during floral senescence? Plant Cell Environ35:217–233

Ruepp A, Zollner A, Maier D, Albermann K, Hani J, Mokrejs M, TetkoI, Guldener U, Mannhaupt G, Munsterkotter M, Mewes H (2004)The FunCat, a functional annotation scheme for systematic clas-sification of proteins from whole genomes. Nucleic Acids Res32:5539–5545

Schiestl F (2005) On the success of a swindle: pollination by deceptionin orchids. Naturwissenschaften 92:255–264

Schiestl F (2010) The evolution of floral scent and insect chemicalcommunication. Ecol Lett 13:643–656

Schiestl F, Ayasse M (2001) Post-pollination emission of a repellentcompound in a sexually deceptive orchid: a new mechanism formaximising reproductive success? Oecologia 126:531–534

Schiestl F, Ayasse M (2002) Do changes in floral odor cause speciationin sexually deceptive orchids? Plant Syst Evol 234:111–119

Schiestl FP, Schlüter PM (2009) Floral isolation, specialized pollina-tion, and pollinator behavior in orchids. Annu Rev Entomol54:425–446

Schiestl F, Ayasse M, Paulus H, Erdmann D, Francke W (1997)Variation of floral scent emission and postpollination changes inindividual flowers of Ophrys sphegodes subsp. sphegodes. JChem Ecol 23:2881–2895

Schiestl F, Ayasse M, Paulus H, Löfstedt C, Hansson B, Ibarra F,Francke W (1999) Orchid pollination by sexual swindle. Nat399:421–422

Schiestl F, Ayasse M, Paulus H, Löfstedt C, Hansson B, Ibarra F,Francke W (2000) Sex pheromone mimicry in the early spiderorchid (Ophrys sphegodes): patterns of hydrocarbons as the keymechanism for pollination by sexual deception. J Comp Physiol A186:567–574

Schlüter P, Xu S, Gagliardini V, Whittle E, Shanklin J, Grossniklaus U,Schiestl F (2011) Stearoyl-acyl carrier protein desaturases areassociated with floral isolation in sexually deceptive orchids.PNAS USA 108:5696–5701

Schrader (1800) Ophrys fusca Link. J Bot 2:324

Funct Integr Genomics

Author's personal copy

Sebastiana M, Figueiredo A, Acioli B, Sousa L, Pessoa F, Baldé A,Pais M (2009) Identification of plant genes involved on the initialcontact between ectomycorrhizal symbionts (Castanea sativa—European chestnut and Pisolithus tinctorius). Eur J Soil Biol45:275–282

Shahri W, Tahir I (2010) Effect of cycloheximide on senescence andpost-harvest performance of Ranunculus asiaticus L. flowers. PakJ Bot 42(5):3577–3585

Shahri W, Tahir I (2011a) Flower senescence—strategies and someassociated events. Bot Rev 77:152–184

Shahri W, Tahir I (2011b) An effective storage protocol for improvingthe postharvest performance in cut spikes of Consolida ajacis.Nieuwl cv. Violet blue. Sci Hortic 129:154–158

Shena M (2002) In: Hans L (ed) Microarray analysis. Wiley, Hoboken,pp 1–25

Shi L, Bielawski J, Mu J, Dong H, Teng C, Zhang J, Yang X,Tomishige N, Hanada K, Hannun Y, Zuo J (2007) Involvementof sphingoid bases in mediating reactive oxygen intermediateproduction and programmed cell death in Arabidopsis. Cell Res17:1030–1040

Stökl J, Paulus H, Dafni A, Schulz C, Francke W, Ayasse M (2005)Pollinator attracting odour signals in sexually deceptive orchids ofthe Ophrys fusca group. Plant Syst Evol 254:105–120

Stökl J, Twele R, Erdmann D, Francke W, Ayasse M (2007)Comparison of the flower scent of the sexually deceptive orchidOphrys iricolor and the female sex pheromone of its pollinatorAndrena morio. Chemoecology 17:231–233

Stökl J, Schlüter P, Stuessy T, Paulus H, Assum G, Ayasse M (2008)Scent variation and hybridization cause the displacement of asexually deceptive orchid species. Am J Bot 95(4):472–481

Stökl J, Schlüter P, Stuessy T, Paulus H, Fraberger R, Erdmann D, SchulzC, Francke W, Assum G, Ayasse M (2009) Speciation in sexuallydeceptive orchids: pollinator-driven selectionmaintains discrete odourphenotypes in hybridizing species. Biol J Linn Soc 98:439–451

Strack D, Busch E, Klein E (1989) Anthocyanin patterns in Europeanorchids and their taxonomic and phylogenetic relevance.Phytochem 28:2127–2139

Suzek B, Huang H, McGarvey P, Mazumder R, Wu C (2007) UniRef:comprehensive and non-redundant UniProt reference clusters.Bioinformatics 23:1282–1288

Tsai W, Hsiao Y, Lee S, Tung C, Wang D, Wang H, Chen W, Chen H(2006) Expression analysis of the ESTs derived from the flowerbuds of Phalaenopsis equestris. Plant Sci 170:426–432

van der Meijden E (1996) Plant defence, an evolutionary dilemma.Contrasting effects of (specialist and generalist) herbivores andnatural enemies. Entomol Exp Appl 80:307–310

van Doorn W (2001) Categories of petal senescence and abscission: are-evaluation. Ann Bot 87:447–456

van Doorn W, Woltering E (2008) Physiology and molecular biologyof petal senescence. J Exp Bot 59:453–480

van Doorn W, Balk P, van Houwelingen A, Hoeberichts F, Hall R,Vorst O, van der Schoot C, van Wordragen M (2003) Geneexpression during anthesis and senescence in Iris flowers. PlantMol Biol 53:845–863

Vandesompele J, De Preter K, Pattyn F, Poppe B, Van Roy N, DePaepe A, Speleman F (2002) Accurate normalization of real-timequantitative RT-PCR data by geometric averaging of multipleinternal control genes. Genome Biol 3(7):RESEARCH0034

Vereecken N, Cozzolino S, Schiestl F (2010) Hybrid floral scentnovelty drives pollinator shift in sexually deceptive orchids.BMC Evol Biol 10:103

Wagstaff C, Bramke I, Breeze E, Thornber S, Harrison E, Thomas B,Buchanan-Wollaston V, Stead T, Rogers H (2010) A specificgroup of genes respond to cold dehydration stress in cutAlstroemeria flowers whereas ambient dehydration stress accel-erates developmental senescence expression patterns. J Exp Bot61:2905–2921

Walton N, Woolhouse H (1986) Enzymes of serine and glycine me-tabolism in leaves and nonphotosynthetic tissues of Pisum sati-vum L. Planta 167:119–128

Wan C, Wilkins T (1994) A modified hot borate method significantlyenhances the yield of high-quality RNA from cotton (Gossypiumhirsutum L.). Anal Biochem 223:7–12

Werner A, Romeis T, Witte C (2010) Ureide catabolism in Arabidopsisthaliana and Escherichia coli. Nat Chem Biol 6:19–21

Woodson W, Park K, Drory A, Larsen P, Wang H (1992) Expression ofethylene biosynthetic pathway transcripts in senescing carnationflowers. Plant Physiol 99(2):526–532

Xu Y, Hanson M (2000) Programmed cell death during pollinationinduced petal senescence in petunia. Plant Physiol 122(4):1323–1333

Xu S, Schlüter P, Scopece G, Breitkopf H, Gross K, Cozzolino S,Schiestl F (2011) Floral isolation is the main reproductive barrieramong closely related sexually deceptive orchids. Evol 65:2606–2620

Yu L, Setter T (2003) Comparative transcriptional profiling of placentaand endosperm in developing maize kernels in response to waterdeficit. Plant Physiol 131:568–582

Zhou L, Wang Y, Peng Z (2011) Molecular characterization andexpression analysis of chalcone synthase gene during flowerdevelopment in tree peony (Paeonia suffruticosa). Afr JBiotechnol 10:1275–1284

Funct Integr Genomics

Author's personal copy