Analysis of gene expression during the transition to climacteric phase in carnation flowers (Dianthus caryophyllus L.)

It has been generally thought that in ethylene-sensitive plants such as carnations, senescence proceeds irreversibly once the tissues have entered the climacteric phase. While pre-climacteric petal tissues have a lower sensitivity to ethylene, these tissues are converted to the climacteric phase at a critical point during flower development. In this study, it is demonstrated that the senescence process initiated by exogenous ethylene is reversible in carnation petals. Petals treated with ethylene for 12h showed sustained inrolling and senescence, while petals treated with ethylene for 10h showed inrolling followed by recovery from inrolling. Reverse transcription–PCR analysis revealed differential expression of genes involved in ethylene biosynthesis and ethylene signalling between 10h and 12h ethylene treatment. Ethylene treatment at or beyond 12h (threshold time) decreased the mRNA levels of the receptor genes (DcETR1, DcERS1, and DcERS2) and DcCTR genes, and increased the ethylene biosynthesis genes DcACS1 and DcACO1. In contrast, ethylene treatment under the threshold time caused a transient decrease in the receptor genes and DcCTR genes, and a transient increase in DcACS1 and DcACO1. Sustained DcACS1 accumulation is correlated with decreases in DcCTR genes and increase in DcEIL3 and indicates that tissues have entered the climacteric phase and that senescence proceeds irreversibly. Inhibition of ACS (1-aminocyclopropane-1-carboxylic acid synthase) prior to 12h ethylene exposure was not able to prevent reduction in transcripts of DcCTR genes, yet suppressed transcript of DcACS1 and DcACO1. This leads to the recovery from inrolling of the petals, indicating that DcACS1 may act as a signalling molecule in senescence of flowers.


Introduction
Petal senescence is the final event in floral development and is mediated by a sequence of highly controlled physiological and biochemical changes that are regulated by ethylene in many flower species (Halevy and Mayak, 1981;van Doorn and van Meeteren, 2003).In ethylene-sensitive flowers such as carnations, ethylene perception is an indispensable requirement to initiate and sustain the ethylene-mediated senescence programme (Thomas et al., 1985;Borochov and Woodson, 1989).When ethylene is perceived by the receptors, the ethylene signal is sent through a sequence of biochemical events that regulate the expression of ethylene-responsive genes leading to ethylene biosynthesis and ultimately senescence of flowers.The ability to perceive or respond to ethylene is most probably mediated by changes in ethylene signalling during flower development (Halevy and Mayak, 1981;Woodson and Lawton, 1988;Verlinden et al., 2002).
Carnation flowers show characteristic floral senescence patterns in response to ethylene.An inward rolling of the petals (petal inrolling) occurs shortly after exposure to ethylene, prior to the initiation of petal wilting; this is regarded as an initial morphological indicator of senescence in carnations (Halevy and Mayak, 1981;Kim et al., 1998;Reid and Çelikel, 2008).Carnation petal senescence appears to be under the control of autocatalytic ethylene biosynthesis as it coincides with the climacteric rise in ethylene synthesis (Halevy and Mayak, 1981;Wang and Woodson, 1989).
Ethylene biosynthesis in plants is under stringent metabolic regulation during development and senescence.The conversion of S-adenosylmethionine (SAM) to 1-aminocyclopropane-1-carboxylic acid (ACC) by ACC synthase (ACS) members is the first rate-limiting step in the ethylene biosynthesis pathway.Subsequent to this conversion, ACC is converted to ethylene by ACC oxidase (ACO) members (Yang and Hoffman, 1984;Kende, 1993;Schaller and Kieber, 2002).It has been proposed that there are two different systems of ethylene synthesis: system 1 is responsible for the basal low level of ethylene production (pre-climacteric phase), detectable in all plant organs, and functions during plant growth and development; and system 2 mediates autocatalytic ethylene production (climacteric phase) and functions during flower senescence and fruit ripening (McMurchie et al., 1972;Yang and Hoffman, 1984;Kende, 1993).Transition of tissues to the climacteric phase during floral senescence in carnation has been shown to be accompanied by changes in transcripts of ethylene biosynthesis genes (Woodson et al., 1992;ten Have and Woltering, 1997;Verlinden et al., 2002;Jones, 2003).
Autocatalytic ethylene synthesis during petal senescence occurs as a result of changes in the ethylene sensitivity of the tissues, and the exposure of pre-climacteric petals to exogenous ethylene can lead to transition to autocatalysis of ethylene synthesis (Mayak and Kofranek, 1976;Halevy and Mayak, 1981;Wang and Woodson, 1989).Previous work has concluded that once petal tissues have entered into this autocatalytic synthesis of ethylene, the senescence process is generally irreversible (Halevy and Mayak, 1981;Yang and Hoffman, 1984).However, in a previous study of carnations (Dianthus caryophyllus L. 'Glacier'), petals exhibited a reversible senescence process.
In this work, the molecular events and ethylene levels that are associated with the climacteric response were studied.Pre-climacteric petal tissues have a lower sensitivity to ethylene and, at a critical point, this shifts the tissues to start autocatalytic ethylene synthesis.The relationship between the change in ethylene sensitivity and senescence of petals was characterized by monitoring petal inrolling and recovery patterns in response to different doses of ethylene.A molecular approach was also taken to explore the genes involved in the transition from the pre-climacteric to climacteric phase.As part of this approach, the mRNA levels of genes involved in ethylene biosynthesis and signalling were monitored, to test the hypothesis that specific genes in ethylene signalling may suppress the accumulation of the ethylene biosynthesis genes, and inactivation of the genes by constitutive perception of ethylene may induce accumulation of the biosynthesis genes.

Plant materials
Rooted cuttings of carnation plants (D. caryophyllus L. cv.Glacier) were planted in 15 cm plastic pots in growing medium (Metro-Mix Special Blend, SUNGRO Horticulture Distribution Inc., Bellevue, WA, USA).Plants were grown in a greenhouse at 22/16 °C day/night temperature and drip-irrigated every other day with half-strength Hoagland nutrient solutions.Supplementary lighting [220 µmol m -2 s -l photosynthetic photon flux density (PPFD) at plant level] was provided by high-pressure sodium lamps (Philips 600 W Master GreenPower) to ensure a photoperiod of 16 h.These lamps were turned on automatically when the intensity of natural lighting was <350 µmol m -2 s -l .Plants were pinched 4 weeks after planting and transplanted into 30 cm plastic pots.After harvest, carnation flowers were immediately placed in tap water for all experiments.

Kinetic analysis for inrolling and angle of petals
The outer whorl of petals was detached from fully open flowers by carefully pulling them out from the receptacles and the individual petals were immediately placed in 1.5 ml micro tubes containing distilled water.To measure petal inrolling due to ethylene, petals were enclosed in a transparent plastic chamber (25 litres) with air circulation by a small fan at 21 °C under light conditions (10 µmol m -2 s -1 ).Ethylene was injected into the chamber to give a final concentration of 10 µl l -1 and withdrawn from the treatment chamber after 10 h of incubation.During incubation of the petals, the front and side of the petals were photographed with a digital camera at 1 h intervals for 6 d.The diameter and angle of petals were analysed from the images using the software ImageJ (http://rsb.info.nih.gov/ij).To compile graphs showing the changes in petal angle, the angle was defined as 0 ° when the petal stands vertically, -90 ° when the petal is perpendicular to 0 ° in an anticlockwise direction, and 90 ° when the petal is perpendicular to 0 ° in a clockwise direction.

Ethylene treatment and measurement of petal diameter
To determine the effect of ethylene treatment time on the inrolling, petals were enclosed in a plastic chamber (117 litres) with air circulation by a small fan at 21 °C under dark conditions.Ethylene was injected into the chambers to give a final concentration of 10 µl l -1 and the petals were incubated under ethylene atmosphere for various durations (9, 10, 11, and 12 h).For treatment with various doses of ethylene, petals were incubated in the treatment chambers with 1, 10, or 100 µl l -1 ethylene for 10 h.
For ethylene treatment on the petals at different stages, the outer petals were detached from carnation flowers at three distinct stages of flower development: day 1 (D1), flowers beginning to open; day 2 (D2), flowers fully opened; and day 3 (D3), flowers fully opened with appearance of the stigma (Fig. 1A).The stages of the petals were defined as 45 °, 90 °, and 120 ° based on the degree of the angles of the outer petals with respect to the axis of the pedicle (Fig. 1A).The petals in three different stages were incubated in the treatment chamber with 10 µl l -1 ethylene for 10 h.Untreated petals were incubated in the same chambers with normal air and used as control.
Ethylene treatment was terminated by transferring the petals from the treatment chambers to normal atmospheric air.After the treatments, the petals were held in the laboratory environment at 21 ± 2 °C, 30 ± 10% relative humidity, and ambient laboratory fluorescent lighting at ~10 µmol m -2 s -1 .Petal diameter was measured using digital calipers before ethylene treatment (-12 h) and at 0 (20 min in air), 4, 8, 12, 16, 24, and 48 h following ethylene removal.

Treatment with 1-methylcyclopropene (1-MCP)
The ethylene binding antagonist, 1-MCP was applied via SmartFresh (AgroFresh Inc., Philadelphia, PA, USA) tablets in the treatment chamber at 21 °C under dark conditions.To generate 100 nl l -1 1-MCP gas, a grey research tablet (0.026 mg 1-MCP) and an activator tablet was added to a beaker containing 18 ml of activator solution in the chamber.The tubes with petals were transferred to the chamber along with a small fan to provide air circulation during the treatment and then the chamber was tightly sealed for 8 h.After the treatments, the petals were held in the laboratory environment.For 1-MCP treatment following 10 µl l -1 ethylene exposure, carnation petals (90 °) were treated with 100 nl l -1 1-MCP for 8 h and kept under normal atmospheric air.

Inhibition of ethylene biosynthesis
Carnation petals (90 ° stage) were treated either simultaneously with aminoethoxy-vinylglycine (AVG; Sigma-Aldrich Co., USA) and ethylene or with AVG prior to ethylene.For the simultaneous treatment, the petals were placed in the tubes containing 100 µM AVG solution and simultaneously incubated in the treatment chamber with 10 µl l -1 ethylene for 12 h.AVG treatments prior to ethylene began 8 h before ethylene exposure.After ethylene treatments, the petals were subsequently treated with 100 nl l -1 1-MCP or air for 8 h.Once AVG treatment started, all petals were kept in the AVG solution throughout the experiment.
Treatments with AVG and ACBC (1-aminocyclobutane-1-carboxylate) (Tocris Bioscience, UK), an inhibitor of ACC oxidase, were carried out before ethylene exposure.The petals were held in tubes containing 100 µM AVG or a mixture of 100 µM AVG plus 100 µM ACBC for 12 h and subsequently treated with 10 µl l -1 ethylene for 12, 14, or 20 h.

Ethylene measurements
Three carnation petals in each treatment were selected and individual petals were enclosed in 25 ml glass vials for 1 h at 21 °C.Gas samples (1 ml) were collected with a gas-tight hypodermic syringe through a rubber septum and analysed for ethylene by gas chromatography (Model 8500, Perkin Elmer Corp., Norwalk, CT, USA) equipped with an aluminium column and flame ionization detector.The detector and oven temperature of the gas chromatograph were held 170 °C and 120 °C, respectively.N 2 was used as a carrier gas at a flow rate of 20 ml min -1 .Each experiment was repeated at least three times.
cDNA synthesis and semi-quantitative RT-PCR Total RNA was isolated from various floral organs using the Trizol method according to the manufacturer's procedure with slight modifications.Petals were collected before ethylene treatment (-12 h) and at 0 (20 min in air), 8, 24, and 48 h following ethylene removal and immediately frozen in liquid N 2 and stored at -80 °C until RNA isolation.Individual petals (~300 µg) were ground in the presence of liquid nitrogen to a fine powder using a pre-chilled mortar and pestle and homogenized in 1 ml of Trizol reagent (Invitrogen, Carlsbad, CA, USA).These samples were separated into aqueous and hydrophobic phases with 200 µl of chloroform by centrifugation.The RNA was precipitated from the aqueous phase with 50 µl of 5 M NaCl and 0.5 ml of isopropanol, washed with 80% ethanol, and resuspended in RNasefree H 2 O.Total RNA was quantified with a NanoDrop DN-1000 spectrophotometer (NanoDrop Technologies Inc., Wilmington, DE, USA).RNA samples were treated with RNase-free DNase prior to RT-PCR, and first-strand cDNA was synthesized from 2 µg of total RNA with 1 µg of oligo(dT) 18 primer, dNTPs, RNA inhibitor, buffer, and M-MLV reverse transcriptase in a final volume of 25 µl according to the manufacturer's instructions (Promega, Madison, WI, USA).The reverse transcription was performed in a PTC-200 PCR machine (MJ Research Inc., MA, USA) with the following temperature parameters: 15 min at 70 °C followed by 1 h at 42 °C.
Gene-specific primers were designed for the ethylene biosynthesis genes (DcACS1, DcACS2, DcACS3, and DcACO1) and ethylene signalling genes (DcETR1,DcERS1,DcERS2,DcCTR1,DcCTR2,DcEIL1,DcEIL2,DcEIL3,and DcEIL4) and synthesized by Integrated DNA Technologies (Coralville, IA, USA).Carnation actin (DcACT1) was used as an internal control.Primer pairs used for semi-quantitative PCR analysis are listed in Table 1.For PCRs, 1 µl of cDNA was used as a template with 2 µl of PCR buffer, 0.2 µl of Taq polymerase, dNTPs, and forward and reverse primers in a final volume of 20 µl.PCR amplification was performed with the following temperature parameters: 95 °C for 3 min and then cycled 35 or 40 times at 95 °C for 20 s, 60 °C for 30 s, and 72 °C for 40 s, with a final extension of 7 min at 72 °C.The PCR products were then analysed using a 1% (w/v) agarose gel and stained with ethidium bromide.The gels were visualized under UV light and the images were taken using a gel documentation system (Gel Dec XR+, Bio-Rad, Hercules, CA, USA) and quantified with ImageJ 1.43u (National Institutes of Health, Bethesda, MD, USA).The relative level of the band was shown as the absolute integrated absorbancy normalized to the relative actin band.All experiments were performed with three independent biological replicates.

Phylogenetic analysis
Protein sequences for the carnation EIL family (DcEIL1-DcEIL4), and Arabidopsis EIN3 (AtEIN3) and EIL1 (AtEIL1) were obtained from the GenBank database (NCBI) and aligned by using ClustalX (version 1.83).Based on the protein sequence similarities, an unrooted Neighbor-Joining phylogenetic tree was constructed by using MEGA software (version 5.05).

Petal inrolling petal angle and recovery patterns
In carnation 'Glacier' flowers, three distinct stages can be distinguished during flower opening (D1, D2, and D3) and the outer petals show 45 °, 90 °, and 120 ° angles, respectively (Fig. 1A).The fully opened flowers (D3) at 3 d after harvest show turgid and fully expanded petals, while the senescent flowers (D10) 10 d after harvest are completely wilted and display inrolled petals (Fig. 1B).Ethylene production in whole flowers rapidly increases at the later stages and peaks at day 9 (Fig. 1C).DcACO1 expression in petals shows a similar pattern to the ethylene production in whole flowers.DcACO1 transcript greatly increases on day 9 when the petal diameter starts to decrease, and then rapidly decreases coincident with a complete inrolling of petals (Fig. 1D, E).The onset of petal inrolling is accompanied by a significant increase in DcACO1 and ethylene synthesis, indicating that petal inrolling is an excellent morphological indicator of senescence in carnation 'Glacier'.
To determine recovery patterns associated with petal inrolling, petals were exposed to 10 µl l -1 ethylene for 10 h, and the changes in petal diameter and angle were measured.The kinetic analysis revealed that petals have three phases of development associated with ethylene responses: phase 1, delay in response; phase 2, rapid inrolling; and phase 3, recovery (Fig. 2A, B).Prior to ethylene exposure, petal diameter and angle are ~38 mm and -130 °, respectively.In phase 1, petal diameter changed minimally with ethylene exposure.In phase 2, the petals rapidly inrolled after the withdrawal of ethylene from the treatment chamber, losing ~70% of their initial diameter (11 mm) within 8 h; thus displaying complete inrolling of the petals.In phase 3, recovery from the inrolling was initiated 8 h after ethylene removal and petals recovered their initial diameter within 48 h.
Interestingly, the petal angle displayed changes earlier than inrolling in the presence of ethylene and rapidly changed vertically following the removal of ethylene.Petal angle changed ~190 ° from the initial angle (-130 °) within 8 h, and rarely recovered with time.Since no correlation was seen with ethylene production, it was hypothesized that the change in petal angle is most probably related to floral opening, and that this developmental process is irreversible in carnation petals.
In order to determine whether the rapid response in petal inrolling results from the transition to air, petals were treated with ethylene and transferred to air after 10 h and 12 h treatment, respectively (Fig. 2C).While petal diameter was not changed during ethylene treatment, it rapidly decreased after ethylene removal after 10 h and 12 h of ethylene treatment.The petals that were treated with ethylene for 10 h lost 40% of their initial diameter at 2 h after ethylene removal (12 h) and recovered after 48 h.The petals treated with ethylene for 12 h retained their initial diameter at 12 h, but then their diameter continued to decrease quickly and did not recover.This result shows that the petal inrolling (diameter) response is initially delayed in the presence of ethylene and occurs immediately after the removal of exogenous ethylene.

Relationship between petal inrolling and ethylene sensitivity
To determine the threshold time of ethylene perception for petal inrolling, petals were treated with 10 µl l -1 ethylene for 9, 10, 11, and 12 h and transferred to air atmosphere at time 0. As shown in Fig. 3A, the petals treated for 9 h with ethylene showed a slight decrease in petal diameter following the transfer to air.The petals treated for 10 h with ethylene lost ~50% of their initial diameter at 8 h, indicating a complete inrolling.However, the petals began to recover from the inrolling following the transfer to air and reverted to their initial diameter 40 h later.On the other hand, Petal angle was defined as 0 ° when the petal stands vertically, -90 ° when the petal is perpendicular to 0 ° in an anticlockwise direction, and 90 ° when the petal is perpendicular to 0 ° in a clockwise direction.(C) Petals were incubated in the treatment chamber with 10 µl l -1 ethylene and the ethylene gas was removed from the chamber after 10 h and the petals treated with ethylene for 11 h and 12 h fully inrolled at 8 h and showed a complete senescence of petals due to inrolling, wilting, and browning within 16 h (Fig. 3A).
The concentration of ethylene also significantly affected petal inrolling (Fig. 3B), and, following ethylene treatments, petals lost ~20, 40, and 65% of their initial diameter within 8 h after treatment with 1, 10, and 100 µl l -1 ethylene, respectively.While the petals in 1 µl l -1 and 10 µl l -1 ethylene returned to their initial diameter within 48 h after removal from ethylene, the petals in 100 µl l -1 ethylene failed to recover.On the other hand, there were no significant differences in petal diameter between the petals at different developmental stages.All petals showed similar inrolling at 8 h, followed by the recovery from the inrolling (Fig. 3B).These results indicate that the recovery from petal inrolling depends on the duration and concentration of ethylene treatment rather than the stage of floral development, and that ethylene treatment lasting longer than the threshold time (12 h) is required to elicit the sustained progression of petal senescence.
To demonstrate whether the progression of petal inrolling is caused by ethylene perception, carnation petals were treated with 1-MCP, an inhibitor of ethylene binding that has been shown to bind competitively to the ethylene receptors with a much higher affinity for the receptors than that of ethylene Fig. 3. Change in petal diameter in response to ethylene and 1-MCP.Petals were (A) treated with 10 µl l -1 ethylene (E) for various times (9-12 h) and (B) treated with various concentration of ethylene (1, 10, or 100 µl l -1 ) for 10 h.(C) Petals in three different stages (45, 90, and 120 °) were treated with 10 µl l -1 ethylene for 10 h.(D) Effect of 1-MCP on change in petal diameter.Petals were treated with 1-MCP (E+M) for 8 h following ethylene treatment (E) for either 10, 11, or 12 h.The control is air-treated petals.Petal diameter was measured before ethylene treatment (-12 h) and at 0 (after 20 min in air), 4, 8, 12, 16, 24, and 48 h after ethylene removal.Data represent the mean ±SE of five replicates.(Sisler et al., 1996;Hall et al., 2000;Binder et al., 2004).Petals were treated with ethylene for 10, 11, and 12 h and subsequently treated with 1-MCP.1-MCP treatment following ethylene exposure slightly suppressed petal inrolling in the petals treated with ethylene for 11 h, but was not able to disrupt the progress of petal inrolling in the petals treated with ethylene for 12 h (Fig. 3D).This suggests that once the petal tissues have entered the climacteric phase, blocking of ethylene is not able to suppress the progression of petal senescence in carnations.

Changes in the mRNA levels of ethylene biosynthesis genes during petal inrolling and recovery
To determine whether the transcript levels of ethylene biosynthesis genes differed between petals that were inrolled or had recovered, RT-PCR analysis was conducted on petals treated with 10 µl l -1 ethylene for 10 h (E10-petals) or 12 h (E12-petals).While DcACS1 transcript was undetectable in control petals, it greatly increased subsequent to ethylene treatment (Fig. 4A).After transfer of the petals to ethylenefree air, DcACS1 gene transcript in E10-petals decreased rapidly and was undetectable at 24 h.In contrast, the DcACS1 transcript remained high in E12-petals for at least 48 h.Although DcACS2 generally showed lower expression than DcACS1, the expression pattern of DcACS2 was similar to that of DcACS1 in air and ethylene (Fig. 4B).On the other hand, DcACS3 transcript levels were initially high, but declined for the first 12h.This occurred whether or not exogenous ethylene was applied and might be due to wounding and subsequent low levels of ethylene (Fig. 4C).In summary, differential regulation of the three ACS genes was observed in carnation during the senescence programme.
Additional biosynthesis genes were analysed to determine if changes were associated with ethylene treatment.DcACO1 expression mirrored the expression patterns of DcACS1.Specifically, after ethylene treatment and subsequent transfer to air, DcACO1 levels in E10-petals quickly decreased, while levels in E12-petals remained high (Fig. 4D).mRNA levels of DcSAMS1 (SAM synthetase) were also measured and, while transcript was abundant in control petal tissues, it declined rapidly upon ethylene treatment.DcSAMS1 expression returned to pre-treatment levels in E10-petals, but not in E12-petals (Supplementary Fig. S1 available at JXB online).However, overall ethylene production increased dramatically subsequent to ethylene treatment, and in E10-petals peaked at 8 h before declining; whereas in E12-petals ethylene production peaked at 8 h and was followed by a second dramatic increase at 24 h (Fig. 4E).The second peak in ethylene synthesis was 2-fold higher than the first peak, and it was hypothesized that this represents the transition to climacteric phase.This parallels the morphological observations as E12-petals showed a steady decrease in petal diameter and a complete wilting at 24 h associated with this second ethylene peak.In contrast, E10-petals showed a recovery in diameter at 24 h simultaneously with the drop in ethylene synthesis (Fig. 4E, F).
The results show that the transcript levels of DcACS1, DcACS2, and DcACO1 have similar patterns of regulation by ethylene and that changes in these genes generally correlate with changes in ethylene biosynthesis.When transcript levels decrease, ethylene levels decrease and petal recovery can occur.In contrast, when the levels of these genes remain elevated, ethylene levels rise and petal senescence occurs.

Changes in the mRNA levels of ethylene receptors and downstream regulators during petal inrolling and recovery
To identify the role of ethylene perception in regulation of petal inrolling and recovery, the transcript abundance of the ethylene receptors (DcETR1, DcERS1, and DcERS2), CTR genes, and EIL genes was determined in petals using RT-PCR.Three previously identified ethylene receptor genes, DcETR1, DcERS1, and DcERS2 (Charng et al., 1997;Shibuya et al., 1998;Nagata et al., 2000) were analysed.While transcript levels of these receptor genes showed some variation in control petals during the time-course of these experiments, these changes were distinct from the 12 h treatment and are likely to be due to wounding.After ethylene treatment for 12 h, gene expression of DcETR1, DcERS1, and DcERS2 decreased.In contrast, 10 h ethylene treatment resulted in transient changes in DcETR1 and DcERS1 transcripts and no change in DcERS2 (Fig. 5).
Downstream targets of the receptors were also analysed including CTR, that functions as a negative regulator of ethylene signalling by forming a complex with the receptors on the endoplasmic reticulum (Kieber et al., 1993;Gao et al., 2003;Huang et al., 2003), and EIN3-Like (EIL) transcription factors predicted to function downstream of CTR1.Application of ethylene for 10 h caused a transient decrease in the transcript levels of both DcCTR1 and DcCTR2, and 12 h treatment with ethylene resulted in a prolonged decrease in the levels of both genes (Fig. 5).These decreases in CTR gene levels are predicted to cause constitutive ethylene signalling for as long as CTR gene level are low.The four EIL transcription factor genes (DcEIL1-DcEIL4) identified in carnations (Waki et al., 2001;Iordachescu and Verlinden, 2005) were also analysed.While DcEIL1 and DcEIL2 (DcEIL1/2) showed relatively inconsistent expression patterns (Fig. 6A), transcript levels of DcEIL3 increased considerably with ethylene treatment.Interestingly, 24 h after ethylene removal, DcEIL3 transcript in E10-petals rapidly dropped to levels similar to those in untreated petals; however, DcEIL3 levels in E12-petals remained high for 48 h.DcEIL4 transcript increased only slightly in E12-petals at 48 h.Increased levels of DcEIL3 have been predicted to lead to ethylene responses (Iordachescu and Verlinden, 2005), and the present results support previous observations.Since EIL1 is phylogenetically and functionally closely related to EIN3, a positive regulator of ethylene perception (Binder et al., 2007), the relationship among EIL and EIN3 family members in carnation and Arabidopsis was examined.A phylogenetic analysis (Neighbor-Joining method) was performed based on predicted protein sequences, and it was determined that DcEIL3 is more closely related to Arabidobsis EIN3 and EIL1 than the other DcEIL genes (Fig. 6B).It is hypothesized that DcEIL3 contributes most to ethylene signalling, consistent with the observation that only DcEIL3 showed an expression pattern that changed in response to ethylene.These results are consistent with previous work showing that DcEIL3 is regulated by ethylene and during flower senescence (Iordachescu and Verlinden, 2005).

Treatment of petals with ethylene synthesis antagonists
To confirm further that ACS may be a main determinant in the transition to climacteric phase, petals were treated with AVG, an inhibitor of ACS activity (Baker et al., 1977;Yang and Hoffman, 1984).When petals were simultaneously treated with AVG and 12 h ethylene (EA), AVG was not able to suppress petal inrolling.However, combined addition of AVG and 1-MCP with 12 h ethylene treatment (EA+M) interrupted the inrolling progress, allowing recovery (Fig. 7A).Transcript levels of DcCTR1 and DcCTR2 were considerably lowered by 12 h ethylene treatment (E), and addition of AVG (EA) was not able to inhibit this decrease in the transcript levels.Alternatively, EA+M effectively prevented the decrease in DcCTR1 and DcCTR2 transcripts at 8 h and 48 h (Fig. 7C).In addition, the ethylene-induced increases in DcACS1 and DcACO1 transcripts were not inhibited by AVG (EA), but were completely suppressed by EA+M (Fig. 7D).These results combined with results shown in Fig. 4 indicate that once ethylene sensitivity has reached a particular threshold level, it requires both 1-MCP to block ethylene perception and AVG to block ethylene synthesis in order to inhibit the progression of senescence.
In contrast, when the petals were treated with AVG prior to ethylene exposure (AE), AVG promoted recovery from inrolling at later stages.Treatment with 1-MCP (AE+M) effectively inhibited the inrolling earlier at 24 h yet demonstrated similar phenotypic responses, as no differences were observed in the final petal diameter at 48 h between AE and AE+M (Fig. 7B).In summary, AE did not inhibit the decrease in DcCTR1 and DcCTR2 in ethylene-treated petals, but suppressed the DcACS1 and DcACO1 transcripts at 8 h and 48 h.AE+M effectively inhibited the reduction in DcCTR1 and DcCTR2 transcripts and suppressed the ethylene-inducible increase in DcACS1 and DcACO1 transcripts (Fig. 7C, D).These data led to the hypothesis that levels of DcCTR1 and DcCTR2 are strongly correlated with DcACS1 and DcACO1 transcripts.These results also suggest that suppression of ACS is able to inhibit DcACS1 and DcACO1 transcripts independently of the inactivation of DcCTR genes.To determine whether the transition to climacteric phase by ethylene can be inhibited by suppression of ACS and ACO activity, petals were simultaneously treated with AVG and ACBC, an inhibitor of ACO activity (Kosugi et al., 1997), for 12 h prior to ethylene exposure.As shown in Fig. 8, petal diameter rapidly decreased by ethylene treatment (E) and the petal inrolling was not inhibited by ACBC single treatment (AcE).In contrast, inrolling induced by 12 h or 14 h ethylene treatment was interrupted by treatments with AVG (AE) or AVG plus ACBC (A+AcE).Petals treated with 12 h ethylene completely recovered from the inrolling at 72 h with AE and at 48 h with A+AcE.Even when the petals were treated with ethylene for 14 h, the petals attained their pre-treatment diameter by 72 h with AE or A+AcE.These results indicate that ACS activity is essential for transition to the climacteric phase and that the suppression of ACS interrupts the senescence programme.

Discussion
The first morphological indication of senescence via petal inrolling occurs subsequent to a surge of ethylene synthesis in the flowers at the climacteric phase.During senescence, petals roll inward as a consequence of asymmetric tissue sensitivity to ethylene, as continuous perception of ethylene results in lipid peroxidation, loss of membrane integrity on the adaxial side of petals, and a subsequent reduction in turgor pressure leading to shrinkage of cells (Thompson, 1988;Kim et al., 1998).
While it has generally been thought that the inrolling of petals during natural flower senescence is an irreversible process (Halevy and Mayak, 1981), some previous studies suggested reversibility of senescence or hormone sensitivity in response to exogenous ethylene in plant tissues (Molisch, 1938;Woltering and Harkema, 1987;Woodson and Lawton, Fig. 5. Expression of ethylene receptor genes, DcETR1, DcERS1, and DCERS2, and the downstream genes of the receptors, DcCTR1 and DcCTR2, in petals.Petals were treated with air only (control), or 10 µl l -1 ethylene for 10 h (E 10 h) or 12 h (E 12 h).The grey boxes indicate the time period when the petals were exposed to ethylene.Petal samples were collected before ethylene treatment (-12 h) and 20 min, 8, 24, and 48 h after ethylene removal.Data represent the mean ±SE of five replicates.
1988; ten Have and Woltering, 1997;Thomas et al., 2003).In this study, unique phases of development were identified and it was determined that this process can be reversible.
Kinetic analyses indicate that petals treated for 10 h with ethylene demonstrate three distinct phases in ethylene responses: phase 1, an initial lag phase during ethylene treatment before morphological changes become evident; phase 2, rapid inrolling of petals after removal from ethylene; and phase 3, the recovery and/or steady state.Interestingly, the petals rarely responded to ethylene in the presence of the hormone (phase 1) and rapidly lost ~50% of their initial diameter within 2 h after the withdrawal of ethylene from the treatment chamber (phase 2).It is hypothesized that the delay in response (phase 1) and rapid transition from phase 1 to 2 might be caused by an inhibitory effect of exogenous ethylene (autoinhibition) resulting from suppression of ACC synthesis and ACS activity (Riov and Yang, 1982;Liu et al., 1985;Philosophhadas et al., 1985;Kende, 1993).In this model, after ethylene removal, ACC becomes available and ACS is active, releasing autoinhibition.Potentially, this leads to an increase in ethylene synthesis, concomitant with rapid inrolling of petals.This hypothesis is supported by previous studies demonstrating that application of exogenous ACC or inhibition of ethylene binding restores ethylene production in the tissues under autoinhibition (Riov and Yang, 1982;Philosophhadas et al., 1985).Once the tissues reach the climacteric phase, exogenous ethylene is no longer inhibitory, as shown in the result of 20 h ethylene treatment in Fig. 8.The present findings combined with previous work on tomato suggest unique roles for the different ACS members within specific tissues during the pre-climacteric and climacteric phase (Barry et al., 2000;Barry and Giovannoni, 2007;Tsuchisaka et al., 2009).
It was observed that recovery from inrolling coincided with increases in receptor transcript levels.This led to the hypothess that both receptors that have lost ethylene due to dissociation as well as the newly synthesized, ethylene-free receptors are active and repressing ethylene signalling.For example, ethylene dissociates with a half-life of 10-12 h from the AtETR1 and AtETR2 receptors, with nearly all the ethylene released after 36 h (Schaller and Bleecker, 1995;O'Malley et al., 2005;McDaniel and Binder, 2012).This timing approximates the recovery pattern of the petals (initial signs of recovery 8 h after ethylene removal and full recovery 40 h later).Thus, recovery from petal inrolling is strongly correlated with the reversion of the receptors to their initial active state, after the dissociation of ethylene.
The results also imply that the suppression of DcACS3 by ethylene is coincident with the activation of DcACS2, and that the accumulation of DcACS2 during the transition period is found together with the sustained accumulation of DcACS1 transcripts, resulting in the initiation of the autocatalytic ethylene production and senescence of petals.This hypothesis is supported by the fact that at the transition point (8 h), low DcACS3 levels and high DcACS2 levels were observed, followed by an increase in the amount of DcACS1 in the E12petals (Fig. 4A-C).There is precedence for this differential regulation of ACS family members in a tissue-specific manner during development, senescence, and fruit ripening (Kende, 1993;Jones and Woodson, 1999a;Barry and Giovannoni, 2007;Tsuchisaka et al., 2009).For example, in tomato plants, LeACS1A and LeACS6 are responsible for system 1 ethylene synthesis, but LeACS4 and LeACS2 function during the transition period and fruit ripening, respectively (Barry et al., 2000;Barry and Giovannoni, 2007).It is hypothesized that DcACS3 could be responsible for system 1 ethylene synthesis which functions during flower development.
Gene expression in response to ethylene has revealed that there is an inverse relationship between the transcript levels of ethylene receptor genes versus DcACS1 and DcACO1.This result supports the previous study that demonstrated a decrease in DcERS2 transcript and a simultaneous increase in ethylene synthesis during senescence of carnation flowers (Shibuya et al., 2002).The decrease in DcETR1, DcERS1, and DcERS2 transcripts was correlated with the reduction in DcCTR1 and DcCTR2 transcripts, and also the induction of the transcription factor DcEIL3.These results are consistent with the model that the receptors are negative regulators of ethylene signalling (Hua and Meyerowitz, 1998;Tieman et al., 2000).Ethylene binding by the receptors suppresses CTR1 action immediately downstream of the receptors, thereby switching on downstream signalling events, resulting in the activation of the positive regulator EIN2 and the subsequent induction of the transcription factor EIN3 (Chao et al., 1997;Alonso et al., 1999;Yanagisawa et al., 2003).In accordance with these interpretations, it is propose that the disappearance of the DcCTR transcripts following 12 h ethylene treatment may represent the turning-off state of CTR action, and this ultimately induces the accumulation of transcription factor DcEIL3, resulting in sustaining of the petal senescence process.
Despite the identification of many key components in ethylene signal transduction, there are still many unanswered questions as to how ethylene signalling is connected to ethylene biosynthesis.It is interesting to note that when DcCTR1 transcript in E10-petals began to recover 24 h after transfer to air, DcEIL3 was completely suppressed simultaneously with the disappearance of DcACS1 and DcACO1.Considering these results, it is believed that this supports the idea that levels of DcCTRs and DcEIL3 contribute to the changes in sensitivity to ethylene in carnation petals.
Mitogen-activated protein kinase (MAPK) modules are also involved in regulating the ethylene biosynthesis pathway.While it is still unclear how the MAPK cascade module is linked to the ethylene signal response pathway, the present data support that DcCTR transcripts are inversely correlated to DcACS1 and DcACO1 transcripts.When ethylene binds to the receptor-CTR1 complex, CTR1 is inactivated; thus no longer blocking downstream signalling events.MPK modules such as MPK3/6 appear to be involved in regulating ethylene biosynthesis by increasing the accumulation of some ACS members; however, the MPK modules are suppressed by CTR1 activation (Ouaked et al., 2003;Christians et al., 2009;Hahn and Harter, 2009).Our work on carnation shows that the accumulation of DcACS1 and DcACO1 is a strict requirement for the transition to climacteric phase and, once the autocatalysis has initiated, DcACS1 and DCACO1 levels remained high.Whether DcCTR1 levels might influence the expression of these two enzymes is unknown.The results demonstrated that inhibition of ACS prior to ethylene exposure was not able to prevent reduction in transcripts of DcCTR genes, yet inhibited transcripts of DcACS1 and subsequently DcACO1.This is also supported by work in Arabidopsis in which the ethylene-constitutive mutant ctr1 has higher levels of ACO2 and produces higher levels of ethylene than the wild type subsequent to ACC application (van Zhong and Burns, 2003;Liu and Zhang, 2004).In the future, the relationship between DcCTR genes and DcACS1 could be more clearly understood by protein analysis.

Fig. 1 .
Fig. 1.Petal inrolling during flower senescence in carnation cv.Glacier.(A) An open flower with fully expanded and turgid petals and a senescent flower with completely inrolled petals.(B) Petals detached from carnation flowers at three distinct stages in flower development: day 1 (D1), flowers beginning to open; day 2 (D2), flowers opened; and day 3 (D3), flowers fully opened with appearance of the stigma.The stages of the petals were defined as 45, 90, and 120 ° based on the degree of petal angles with respect to the axis of the pedicle.(C) Ethylene production by whole flowers during natural senescence.(D) Petal inrolling and relative expression level of DcACO1 during natural senescence of detached petals.(E) Expression profiles of DcACO1 and DcACT1.

Fig. 2 .
Fig.2.Inrolling and recovery patterns of petals.Petals detached from fully open flowers were held in micro tubes containing distilled water and treated with 10 µl l -1 ethylene for 10 h.(A) Changes in petal inrolling and angle were photographed; and (B) petal diameter and angle were analysed from the images.Petal angle was defined as 0 ° when the petal stands vertically, -90 ° when the petal is perpendicular to 0 ° in an anticlockwise direction, and 90 ° when the petal is perpendicular to 0 ° in a clockwise direction.(C) Petals were incubated in the treatment chamber with 10 µl l -1 ethylene and the ethylene gas was removed from the chamber after 10 h and 12 h, respectively.Petal diameter was measured before ethylene treatment (0 h), immediately after ethylene removal (10 h and 12 h), and at 13, 14, 16, 18, and 48 h after ethylene removal.Data represent the mean ±SE of five replicates.

Fig. 4 .
Fig. 4. Expression of ethylene biosynthesis genes.(A) DcACS1, (B) DcACS2, (C) DcACS3, and (D) DcACO1 in petals.(E) Ethylene production by petals.(F) Petal diameter.Petals were treated with air only (control), or 10 µl l -1 ethylene for 10 h (E 10 h) or 12 h (E 12 h).The grey boxes indicate the time period when the petals were exposed to ethylene.Petal samples were collected before ethylene treatment (-12 h) and 20 min, 8, 24, and 48 h after ethylene removal.Data represent the mean ±SE of five replicates.

Fig. 6 .
Fig. 6.Expression of the transcription factors, DcEIL1, DcEIL2 (DcEIL1/2), DcEIL3, and DcEIL4 in petals.(A) Relative expression patterns.Petals were treated with air only (control), or 10 µl l -1 ethylene for 10 h (E 10 h) or 12 h (E 12 h).The grey boxes indicate the time period when the petals were exposed to ethylene.Petal samples were collected before ethylene treatment (-12 h) and 20 min, 8, 24, and 48 h after ethylene removal.(B) Phylogenetic analysis.Protein sequences for the carnation EIL family (DcEIL1-DcEIL4), and Arabidopsis EIN3 (AtEIN3) and EIL1 (AtEIL1) were aligned and a phylogenetic tree was constructed by an unrooted Neighbor-Joining method based on the protein sequence similarities.Data represent the mean ±SE of five replicates.

Fig. 7 .
Fig. 7. Effects of AVG and 1-MCP on changes in petal diameter and gene expression.(A) Petal diameter when the petals were treated simultaneously with 12 h ethylene (E) and AVG (EA) and subsequently with 1-MCP (+M).(B) Petal diameter when the petals were treated with AVG prior to 12 h ethylene (AE) and subsequently 1-MCP.Petals were kept in the AVG solution throughout the experiment once the treatment had started.The grey boxes indicate the time period when the petals were exposed to ethylene.Petal diameter was measured before ethylene treatment (-12 h) and 20 min, 8, 24, and 48 after ethylene removal.The relative expression of (C) DcCTR1 and DcCTR2 and (D) DcACS1 and DcACO1 was detected at 8 h and 48 h after ethylene removal.Data represent the mean ±SE of five replicates.

Fig. 8 .
Fig. 8. Effects of AVG and ACBC on change in petal diameter.Petals were treated with AVG (AE) or AVG+ACBC (A+AcE) before ethylene treatment (E) for either 12, 14, or 20 h.Petals were kept in the AVG or AVG+ACBC solution throughout the experiment once the treatment had started.Petal diameter was measured before ethylene treatment (-20 h) and at 0 h (after 20 min in air) and every 12 h after ethylene removal.Data represents the mean ±SE of five replicates.

Table 1 .
Gene-specific primers used for RT-PCR amplification of cDNA fragments