Introduction

Roses (Rosa hybrida L.) are one of the most important cut flowers among ornamental plants. The longevity of cut rose flowers is a major factor in determining the ornamental value of cut roses. However, the longevity of cut roses is limited due to various factors (Fanourakis et al. 2013;Verdonk et al. 2023). The vase life and senescence characteristics in cut roses depend on cultivars and seasonal variation (In et al. 2017;In and Lim 2018;Macnish et al. 2010). The vase life of cut roses is typically terminated by petal or leaf wilting, discoloration of petals, browning of petal edges, petal or leaf abscission, bending of the peduncle (bent neck), or disease infection (e.g. gray mold disease caused by Botrytis cinerea) (Bergmann and Dole 2020;Ha et al. 2022;Halevy and Mayak 1979;Hammer and Evensen 1991;In et al. 2017;In and Lim 2018). Bending of the peduncle is thought to be caused by vascular occlusions of the flower xylem, impeding water absorption of cut flowers (Bleeksma and van Doorn 2003). This senescence symptom serves as a primary and early factor contributing to the shortened vase life in rose cultivars sensitive to water stress (Lear et al. 2022;Zieslin et al. 1978;). Although the association between bacterial contamination and bent neck is widely acknowledged (Bleeksma and van Doorn 2003;van Doorn et al. 1995;van Doorn 2012), there still exists variation in the incidence rate of bent neck among rose cultivars and across different harvest seasons (In et al. 2016;In and Lim 2018;Lear et al. 2022;Pompodakis et al. 2005). Factors involved in the bent neck are not well identified in cut flowers. Generally, the bent neck of cut flowers is correlated with poor pectin gelation and lignification mechanisms due to low expression levels of pectin or lignin biosynthesis genes during the development and elongation of the flower stems (Bethke et al. 2016;Lear et al. 2022;Perik et al. 2012). The previous study also showed that a bent neck in rose flowers was related to the up-regulation of galactose metabolism and down-regulation of phenylpropanoid biosynthesis genes (Lear et al. 2022). Exposure to ethylene increased the bent neck incidence rate in cut flowers (Mencarelli 1995). However, there have been no recent studies addressing the expression profiles of ethylene, lignin, and pectin-related genes in the process that affects the bent neck in cut roses among growing seasons.

The initiation of floral organ abscission is stimulated by developmental (aging of the flowers, senescence, and cell wall modifications) and environmental signs and plant hormones have been demonstrated to play important roles in regulating floral organ abscission (Estornell et al. 2013;Taylor and Whitelaw 2001). Generally, ethylene is considered the main regulator that accelerates the abscission of the floral organs, while auxin suppresses the floral organ abscission (Meir et al. 2010;Roberts et al. 2002). Exposure to ethylene caused petal abscission in rose flowers, especially in ethylene-sensitive rose cultivars (Ha et al. 2020;In et al. 2017;In and Lim 2018). In rose flowers, the regulation of petal abscission is complex and involves a balance between ethylene and auxin signaling pathways (Gao et al. 2016;Gao et al. 2019). The previous study showed that the ethylene response factor RhERF4 was induced by auxin and directly regulates the expression of RhBGLA1 which encodes a pectin-metabolizing enzyme (Gao et al. 2019). The down-regulation of RhBGLA1 results in less pectin degradation and delayed petal abscission in rose flowers (Gao et al. 2019). A recent finding has shown that an auxin response factor in rose flowers RhARF7 encodes for a signaling protein that can bind to the promoter of the sucrose transporter gene RhSUC2 (Liang et al. 2020). Knockdown or silencing of RhARF7 and/or RhSUC2 results in a reduction in sucrose levels in rose petals and stimulates petal abscission (Liang et al. 2020). Therefore, it is important to identify the variation in expression patterns of these genes across harvest seasons and their relationship to the incidence rate of petal abscission in cut roses.

In this study, we studied the variation in the occurrence of senescence symptoms, gene expression, and vase life of cut roses grown across seasons in the greenhouses. The transcript levels of genes associated with ethylene, pectin, and lignin biosynthesis, auxin response, and sucrose transport were monitored in cut roses on the initial day after the harvest. This aimed to elucidate the impacts of ethylene, lignin, pectin, and auxin on the bent neck and petal abscission of cut roses and to understand how the gene expression and these senescence symptoms are changed by seasons.

Materials and Methods

Plant materials and exportation simulation

The rose ‘S-Pink’ (spray) and ‘Yeonmo’ (standard) (Rosa hybrida L.) flowers were grown in a greenhouse in Jeonju, Republic of Korea. Rose plants were trickle irrigation with a liquid nutrient solution containing 44.93 g L^-1 NH₄NO₃, 17.47 g L^-1 Ca(NO₃)₂‧H₂O, 1.63 g L^-1 KNO₃, 12.04 g L^-1 KH₂PO₄, 27.04 g L^-1 MgSO₄‧7H₂O, and other trace substances. The air temperature, relative air humidity (RH), and light intensity in the greenhouse were recorded at 30-min intervals by using data loggers (WatchDog 1450, Spectrum Technologies, Aurora, IL, USA).

Rose flowers at the half-open stage with outer petals curved were randomly harvested on August 10 (summer) and October 8 (autumn) in 2023. After harvest, cut flowers were immediately placed in buckets containing tap water and transported to the laboratory within 3 h. Cut flowers were held in distilled water and then stored at 8-10°C and relative RH of 50-60% under dark conditions for 3 days for simulation of export conditions. After the transport simulations, the flower stems were recut to a length of 45 cm, including three upper leaves. Each stem of the spray flowers remained five florets. The cut rose flowers were then held in distilled water and a controlled environment room at 25 ± 1°C and RH of 50 ± 2% for senescence symptoms and vase life evaluation.

Postharvest characteristics and vase life evaluation

Variation in postharvest quality of cut flowers between seasons was assessed by measuring changes in relative fresh weight, water uptake, and flower diameter daily. Water balance of cut flowers was calculated by deducting daily transpiration from daily water uptake. The daily transpiration was calculated from the change of fresh weight and water uptake of cut flowers.

The vase life of cut roses was determined as the duration between the day when the cut flowers were placed in the vase and the day when they lost their ornamental value. The cut roses were considered to have reached the end of their vase life when one or more of the following senescence symptoms were observed in flowers: petal wilting (≥ 50% petal or leaf turgor loss), petal abscission, bent neck, discoloration (≥ 50% blue petals), and gray mold disease (GMD) (VBN 2014). Bent neck symptom was descripted as the drying and kinking of the stem just below the flower bud. This symptom was recorded when the flower is at an angle less than 90° and greater than 90° and was classified as two levels, slight and severe bending of neck, respectively. Petal abscission level was evaluated on a relative scale of 1-3 according to the percentage of drop-petal numbers to total petal numbers, as follows: 1, 10-30% of petals dropped; 2, 30-50% of petals dropped; and 3, more than 50% of petals dropped.

RNA isolation

Nine flowers in each cultivar were randomly chosen for sampling for gene expression analysis. Five (spray flowers) or three (standard flowers) petals and pedicel segments (3 cm from flower buds) were collected from cut roses on the first day after harvest. These rose petals and pedicel segments were immediately frozen in liquid nitrogen and then stored at -80°C until RNA extraction. Petals (150 mg) and pedicel segments (300 mg) were ground with liquid nitrogen to a fine powder using a pre-chilled mortar and pestle. Total RNA was isolated from petal and stem tissues using the GeneJET plant RNA purification Mini Kit (Thermo Fisher Scientific Baltics, Vilnius, Lithuania) with slight modifications of the manufacturer’s protocol. Total RNA was quantified at 260/280 nm using a NanoDrop spectrophotometer (NanoDrop Lite Plus, Thermo Fisher Scientific, Waltham, MA, USA).

cDNA synthesis

Purified total RNA was mixed with nuclease-free water and used for cDNA synthesis using a Power cDNA Synthesis Kit (INTRON Biotechnology, Inc., Korea). First strand cDNA was synthesized from 0.1 μg of total RNA using 1 μL of oligo (dT)₁₅ primer, RNase inhibitor, 5xRT buffer, dNTPs, DTT, and AMV RT enzyme in a final volume of 20 μL, according to the manufacturer’s instructions. Reverse transcription was performed in a Bio-Rad PTC-100 Programmable Thermal Controller (MJ Research Inc., Hercules, CA, USA) using the following temperature parameters: 5 min at 75°C followed by 60 min at 42°C.

Quantitative real-time PCR (qRT-PCR)

Gene-specific primers were designed for pectin, lignin, and ethylene biosynthesis, sucrose transport, and auxin response genes and synthesized by Cosmogenetech (Seoul, Korea). Rosa hybrida actin 1 (RhACT1) was used as an internal control to confirm the amount of the template RNA. The sequences of the primers used for qRT-PCR analysis are shown in Table 1. mRNA levels of the target genes were detected using the BIO-RAD CFX Connect Real-Time System (Life Science, Hercules, CA, USA). Reaction mixtures contained 5 μL of cDNA as a template, 1 μL of 10 μM forward and reverse primers, 10 μL of iQ™ SYBR® Green Supermix (Bio-Rad, Life Science, Hercules, CA, USA), and 4 μL sterile distilled water in a final volume of 20 μL dispensed in an optical 96-well plate. The qRT-PCR conditions for the detected genes have been described previously (Ha and In 2023).

Experimental design and data analyses

The vase life experiment followed a completely randomized block design (three replicates with twelve sub-replicates). The qRT-PCR analysis was conducted with nine biological replicates. Data were presented as the means ± standard errors. The differences between the means were tested using t-test at p = 0.05 or p = 0.01. To clarify the correlations between bent neck and petal abscission incidence rates and gene expression in cut roses, Pearson correlation analysis was performed. Statistical analyses were performed using SPSS version 22.0 (IBM Corp., Armonk, NY, USA).

Results

Characteristics of senescence and longevity of cut roses

The incidence rates of bent neck and petal abscission in autumn rose flowers were higher than those of summer flowers in both cultivars (Fig. 1A and B). Significantly, no instances of bent neck or petal abscission were observed in summer flowers of the ‘Yeonmo’ cultivar (Fig. 1A and B). The manifestation of bent neck in cut roses was categorized into two levels: a slight bending (when the flower is at an angle less than 45°; BN45) and the severe bending (when the flower is at an angle greater than 90°; BN90) (Fig. 1C). In ‘S-Pink’, autumn flowers exhibited a markedly higher incidence rate of the BN90 (82.7%) compared to that of summer flowers (4.5%). Notably, the BN45 was recorded in summer flowers (95.5%) (Fig. 1D). Similarly, petal abscission was more severe in autumn flowers than in summer flowers (Fig. 1E). In ‘Yeonmo’, autumn flowers showed a high incidence rate of the BN45 (93.5%) and slight petal abscission (89%) at the end of the vase period (Fig. 1D and E). These results indicate that the occurrence of bent neck and abscission in cut roses may be seasonal dependent.

Petal wilting, discoloration (the petals turn blue), and GMD were also observed in both of the rose cultivars (Fig. 2A and B). Notably, the incidence rate of GMD in autumn flowers was higher than that in summer flowers for both rose cultivars (Fig. 2A and B). The maintenance of the initial fresh weight (IFW) and positive water balance (PWB) of cut flowers differed between seasons (Fig. 2C-F). The time that retaining IFW and PWB for autumn flowers was shorter than that for summer flowers in both rose cultivars (Fig. 2C-F). Consequently, the vase life of summer flowers was longer than that of autumn flowers by 3.3 days in the ‘S-Pink’ cultivar and by 2.2 days in the ‘Yeonmo’ cultivar (Fig. 2G and H). While in this study, the flower opening did not vary between seasons (data not shown).

Gene expression patterns in pedicels

To test the involvement of ethylene, pectin, and lignin in the incidence rate of bent neck in cut roses between seasons, the expression levels of genes related to ethylene (RhACS2 and RhACO1), pectin (RhGAUT1), and lignin (RhC4HL, RhCAD9, RhCOMT1, RhHCT, and RhPRXPX) biosynthesis were detected in pedicels. The selection of these genes for the present experiment was based on previous studies (In et al. 2017;Lear et al. 2022). In ‘S-Pink’, the expression levels of RhACS2 and RhACO1 in pedicels of autumn flowers were higher than those in summer flowers (Fig. 3A). Conversely, in ‘Yeonmo’, RhACS2 expressed similar levels in both seasons, while the relative expression of RhACO1 was higher in autumn flowers than in summer flowers (Fig. 3B).

In contrast to the genes involved in ethylene biosynthesis, the accumulation of pectin biosynthesis-related gene RhGAUT1 in pedicels was lower in autumn flowers than in summer flowers in both rose cultivars (Fig. 3C and D). Among the lignin biosynthesis genes, the expression levels of RhCAD9, RhCOMT1, RhHCT, and RhPRXPX exhibited seasonal variations in both cultivars (Fig. 3E and F). The accumulation of these genes in pedicels was higher in summer flowers than in autumn flowers (Fig. 3E and F). Notably, RhC4HL expression levels did not differ between seasons in both cultivars (Fig. 3E and F). These findings suggest a close association between the incidence rate of bent neck in cut roses and the elevated expression levels of ethylene biosynthesis genes, as well as the diminished expression of pectin and lignin biosynthesis genes in pedicels. The different expression levels of these genes in cut roses may be due to the influence of pre-harvest conditions in the greenhouses.

Gene expression patterns in petals

Previously, the regulation of petal abscission involved a balance between ethylene and auxin signaling pathways. Moreover, auxin regulates the gene that encodes the signaling protein RhARF7, capable of binding to the promoter of the RhSUC2 sucrose transporter. The down-regulation of either RhARF7 or RhSUC2 leads to a decrease in sucrose levels in petals, facilitating petal abscission in cut roses (Liang et al. 2020;Sun et al. 2021). Therefore, to investigate the correlation between the incidence rate of petal abscission and the expression levels of genes associated with ethylene biosynthesis, auxin response, and sucrose transport across harvest seasons, we assessed the mRNA levels of RhACS2, RhACO1, RhARF7, and RhSUC2 in the petals of cut roses. In ‘S-Pink’, the accumulation of RhACS2 in petals was higher in autumn flowers than in summer flowers, while the expression of RhACO1 remained consistent in both seasons (Fig. 4A). In ‘Yeonmo’, the expression of RhACS2 showed no variation between seasons, whereas mRNA levels of RhACO1 were elevated in autumn flowers compared to summer flowers (Fig. 4B). The relative expression of RhARF7 and RhSUC2 was lower in autumn flower than in summer flowers in both rose cultivars (Fig. 4C-F). These data indicate that the high accumulation of ethylene biosynthesis genes and reduced sucrose levels in petals, attributed to the diminished expression of RhARF7 and RhSUC2 in autumn flowers, trigger petal abscission in cut roses.

Relationships between gene expression and senescence symptoms of cut roses

Correlation analysis was conducted to determine the relationships between the incidence rates of bent neck and petal abscission and gene expression in cut roses. Gene expressions of RhACO1 (p < 0.05) (in ‘S-Pink’), RhGAUT1 (p < 0.05), and RhPRXPX (p < 0.01) showed significantly high positive and negative correlations with the incidence rates of bent neck (Fig. 5A). This indicates that higher ethylene level and lower lignin content during the early maturity stage of cut roses increase the incidence rates of bent neck. RhACS2 (in ‘S-Pink’) and RhACO1 (in ‘Yeonmo’) were highly positively correlated with petal abscission incidence rate in cut flowers (r = 0.73 and r = 0.7; p < 0.05; Fig. 5B). This result shows that higher accumulation of these genes in autumn contributes to increasing the incidence rate of petal abscission in cut rose flowers.

Discussion

The senescence characteristics and longevity of cut roses depend on cultivars and seasonal variation (In et al. 2017;In and Lim 2018;Macnish et al. 2010). In this study, the vase life of cut rose flowers was shorter in autumn than in the summer harvest season. Senescence symptoms in cut roses grown in greenhouses also varied among harvest seasons and cultivars. Bent neck and petal abscission were the primary senescence symptoms that terminated the vase life of autumn rose flowers in ‘S-Pink’ and ‘Yeonmo’. Notably, the levels of bent neck and petal abscission in cut roses were more severe in autumn than in summer. Similarly, bent neck and petal abscission also occurred in a considerable number of autumn and winter rose flowers (In and Lim 2018). Previously, the incidence rate of bent neck of greenhouse-grown rose flowers was also found to be related to their lignin and pectin content (Parups and Voisey 1976). It has been demonstrated that the mechanical strength of stems, as determined by the content of lignin and pectin, is correlated with the flower neck’s resistance to bending (van Doorn 1997;Zieslin et al. 1978). The abundance of lignin and pectin in stems is determined by the high accumulation of key genes involved in lignin and pectin biosynthesis (Shin et al. 2021;Xie et al. 2018). In this study, the mRNA levels of genes related to lignin and pectin biosynthesis in summer flowers were higher than in autumn flowers corresponding to the bent neck incidence rates in cut roses.

We found that among lignin and pectin biosynthesis genes, mRNA levels of RhPRXPX and RhGAUT1 were highly negatively correlated with the incidence rates of bent neck in cut roses. PRXPX is the main enzyme that directly converts G unit lignin into conventional lignin in plants (Liu et al. 2018;Xie et al. 2018). Conventional lignin is a complex, irregular polymer composed of phenolic compounds, and its main function is to provide structural support to plant cell walls. It helps to bind cellulose and hemicellulose fibers together, contributing to the overall strength and rigidity of the plant structure (Liu et al. 2018;Xie et al. 2018). GAUT1 encodes a galacturonosyltransferase enzyme involved in the biosynthesis of homogalacturonan, a major component of pectin (Mohnen 2008;Shin et al. 2021). Pectin is an important component of the plant cell wall and plays a role in cell adhesion, cell wall integrity, and communication between cells (Mohnen 2008;Shin et al. 2021). Changes in the expression or activity of GAUT1 can influence the composition and properties of pectin in the cell wall (Shin et al. 2021). Therefore, a higher accumulation of these genes in the pedicels of summer flowers helps cut flowers resistant to bent necks.

A previous study showed that ethylene negatively regulated lignin biosynthesis in Arabidopsis leaves by directly activating the expression of the miR397b/miR857- laccases module. The activation of miR397b/miR857 by ethylene leads to suppression of their target genes, which are mainly involved in the final step of lignin biosynthesis (Gaddam et al. 2022). Therefore, the high expression levels of ethylene biosynthesis genes in autumn rose flowers may contribute to reduced lignin content in pedicels, ultimately resulting in a higher incidence rate of bent neck. However, the regulation of lignin and pectin biosynthesis and metabolism by ethylene is still unclear in rose flowers. Thus, further studies need to be conducted to clarify the involvement of ethylene in the lignin and pectin metabolism as well as the bent neck in rose flowers.

Pectin also can help the cells to absorb and retain water, contributing to the maintenance of turgor pressure in plant cells (Gomes et al. 2010). It has been shown that higher pectin content in stems and petals of cut flowers results in higher relative fresh weight and solution uptake during the vase life of cut flowers (Gomes et al. 2010). Similarly, summer rose flowers with higher expression levels of pectin biosynthesis gene RhGAUT1 had longer periods that maintained initial fresh weight and positive water balance. This might contribute to a longer vase life in summer flowers.

Floral organ abscission often occurs due to a lack of nutrients and competition for carbohydrates/assimilates (van Doorn 2002). In rose flowers, petal abscission is related to a decrease in the transportation of sucrose and auxin to petals (Liang et al. 2020). Auxin plays a positive role in the regulation of sucrose transportation in rose petals through auxin response factor RhARF7 (Liang et al. 2020). Protein RhARF7 binds to the promoter of sucrose transport gene RhSUC2, resulting in an enhanced sucrose transportation to petals (Liang et al. 2020). In this study, the transcript levels of RhSUC2 and RhARF7 in the petals of summer flowers were higher than those of autumn flowers. This may help to maintain a high level of sucrose content in the petals of summer flowers and decrease the petal abscission in cut roses. Additionally, ethylene is known to play a significant role in promoting flower abscission (Dar et al. 2021;van Doorn 2002). The high accumulation of ethylene biosynthesis genes in autumn and winter flowers increased the incidence rate of petal abscission in cut roses (In and Lim 2018). Petal abscission is also related to the degradation of pectin in the abscission zone which is regulated by the balance between ethylene and auxin (Gao et al. 2019). The pectin metabolism in rose petals was modulated by the integration of RhERF1 and RhERF4 and the coordination of ethylene and auxin signals (Gao et al. 2019). In the present study, mRNA levels of RhACS2 and RhACO1 were positively correlated to petal abscission rates in ‘S-Pink’ and ‘Yeonmo’ flowers. This result indicated that higher ethylene sensitivity leads to decreased sucrose transportation to petals and a high level of pectin degradation, resulting in promoting petal abscission in autumn rose flowers. In addition, the high expression levels of ethylene biosynthesis genes in petals also contribute to promoting the susceptibility of autumn flowers to GMD (Ha et al. 2021;Ha et al. 2022), thus decreased vase life of the cut flowers comparing to that of summer flowers.

In conclusion, senescence characteristics in cut roses were found to be dependent on cultivars and growing seasons. The incidence rates of bent neck and petal abscission in summer flowers were significantly lower than those observed in autumn flowers. The higher accumulation of RhACS2 and RhACO1 in autumn may contribute to increased petal abscission and GMD infection, consequently reducing the vase life of cut roses. The incidence rates of bent neck in cut roses were negatively correlated to the mRNA levels of lignin and pectin biosynthesis genes RhPRXPX and RhGAUT1 in pedicels. Our findings are important for further understanding the seasonal changes in molecular characteristics involved in regulating the mechanical strength of pedicels during the bent neck and the petal abscission in cut roses. Thus, breeding techniques and postharvest treatments are potential strategies that could be used to improve lignin and pectin contents in cell wall components and extend the vase life of cut roses.