The Variation of Oncidium Rosy Sunset Flower Volatiles with Daily Rhythm, Flowering Period, and Flower Parts

Oncidium is an important ornamental crop worldwide, and in recent years, the characteristics of the flower aroma have become a concern for breeders. This study used headspace solid-phase microextraction (HS-SPME) and gas chromatography/mass spectrometry (GC-MS) analysis of the volatile compounds to study the aroma characteristics of Onc. Rosy Sunset. A total of 45 compounds were identified, with the major compound being linalool. Onc. Rosy Sunset had the highest odor concentration from 10:00 to 12:00 and lowest from 20:00 to 24:00. The inflorescence emitted the highest quantities of volatile compounds during stages 3–6, which then decreased with the aging of the flowers. In Onc. Rosy Sunset, the sepals and petals were the major parts for the floral fragrance emission, in which linalool content was the highest, whereas the lip and column had a different composition of major volatile compounds, of which benzaldehyde, β-myrcene, and β-caryophyllene dominated.


Introduction
Oncidiinae are native to the Neotropical region, include approximately 87 genera, with 1600 species, varieties or cultivars, and have become economically important ornamental crops worldwide for use in the cut-flower and pot-plant industries [1][2][3]. Floral fragrance is one of the important features of the ornamental orchid, and can improve the aesthetic value, quality of flower products, and their economic merit. Cultivating fragrance is a current trend in orchid breeding, with several studies involving orchid volatile compound analysis and synthesis, mainly focused on Phalaenopsis [4][5][6][7][8]. There are few fragrant cultivars of Oncidium at present, with the minimal research into their aromatic volatile compounds mostly confined to local journals and focused on Onc. Sharry Baby, a large potted plant [9][10][11][12].
Floral fragrance is also a key factor in plant interactions with insects and is important for plant reproduction and evolution [13,14]. The pollinator plays a key role with orchid sexual propagation; therefore, fragrance is especially important for orchids [5].
The extraction of volatile compounds from flowers is an important step in the analysis of floral aroma. The typical methods used include liquid-liquid extraction, purge and trap, steam distillation (SD), simultaneous distillation extraction (SDE), supercritical-fluid extraction (SFE), solid-phase extraction (SPE), and solid-phase microextraction (SPME) [15]. SPME is simple, fast, sensitive, requires no solvent [16], and has been successfully used to analyze volatile compounds in a variety of flowering plants, such as Hosta, Tree peony, Styrax tonkinensis, Narcissus tazetta, Table 2 shows that the volatile components changed with the daily rhythm. The total relative content is expressed in total peak areas. The odor was strongest at 12:00, with a total peak area of 5117, followed by 4555 at 10:00, and it gradually declined until 18:00-04:00, with the lowest values of 535-640, and then rose again at 8:00 with a peak area of 1917. These results were similar to those reported for Phal. Nobby's Pacific Sunset and Gongora bufonia, which showed stronger odor levels at 09:00-13:00 and 10:00, respectively [8,22]. Orchids possess entomophilous flowers, and the aroma volatile components are key for the attraction of insects. Flowers emit different volatile compositions at different times to attract specific pollinators, such as a specific type of bee or diurnal/nocturnal insects [23,24]. The diurnal bee is a pollinator for both Phalaenopsis and Gongora bufonia; therefore, the strongest odor emitted was during the day.  Table 1; x, column, or septum bleed). (A) Original total ion chromatogram; (B) Normalized total ion chromatogram.
There were significant differences in the emission of most single compounds. The exceptions were tiglaldehyde, 2-methyl-2-buten-1-ol, benzaldehyde, β-myrcene, 2-ethylhexanol, methyl salicylate, 2-ethylhexyl acrylate, β-citral, β-caryophyllene, (Z)-β-farnesene, benzyl tiglate, and nerolidol, for which no obvious differences in relative percentage of floral scent emitted over a daily rhythm were found. Linalool had the highest relative percentage change, with a maximum of 37.09-35.8% at 10:00-12:00 and a minimum of 4.72% at 22:00 (Table 2). Such changes were similar to the total relative content changes over the daily rhythm. We note that Borg-Karlson et al. [25] studied linalool as a mate attractant component in the male bee. Not all the volatile compositions peaked during the day, as the relative percentages of β-pinene, limonene, (E)-4,8-dimethyl-l,3,7-nonatriene, menthol, tridecane, and copaene were highest at 18:00-22:00. The emission of cis-linalool oxide, α-terpineol, isopentyl benzoate, and 2-methylpropanoate peaked at nightfall (16:00-18:00; Table 2). Dötterl et al. [23] suggested that changes in aroma components during the daily rhythm of Silene otites may attract specific insects at different times, not only the originally identified moths and mosquitoes. This may explain why some volatile components were highest at nightfall or nighttime for Onc. Rosy Sunset. It is worth noting that (E)-4,8-dimethyl-l,3,7-nonatriene, which is generally emitted from the damaged leaves and attracts insects, is only found in a few flowers, such as Magnolia taxa and Liriodendron tulipifera [26]. In our results, (E)-4,8-dimethyl-l,3,7-nonatriene was emitted in highest concentrations at nighttime, although it is not known whether this is related to the attraction of nocturnal insects. Linalool was a primary component of emission, over all times of the daily rhythm.  Table 1; x, column, or septum bleed). (A) Original total ion chromatogram; (B) Normalized total ion chromatogram.
There were significant differences in the emission of most single compounds. The exceptions were tiglaldehyde, 2-methyl-2-buten-1-ol, benzaldehyde, β-myrcene, 2-ethylhexanol, methyl salicylate, 2-ethylhexyl acrylate, β-citral, β-caryophyllene, (Z)-β-farnesene, benzyl tiglate, and nerolidol, for which no obvious differences in relative percentage of floral scent emitted over a daily rhythm were found. Linalool had the highest relative percentage change, with a maximum of 37.09-35.8% at 10:00-12:00 and a minimum of 4.72% at 22:00 (Table 2). Such changes were similar to the total relative content changes over the daily rhythm. We note that Borg-Karlson et al. [25] studied linalool as a mate attractant component in the male bee. Not all the volatile compositions peaked during the day, as the relative percentages of β-pinene, limonene, (E)-4,8-dimethyl-l,3,7-nonatriene, menthol, tridecane, and copaene were highest at 18:00-22:00. The emission of cis-linalool oxide, α-terpineol, isopentyl benzoate, and 2-methylpropanoate peaked at nightfall (16:00-18:00; Table 2). Dötterl et al. [23] suggested that changes in aroma components during the daily rhythm of Silene otites may attract specific insects at different times, not only the originally identified moths and mosquitoes. This may explain why some volatile components were highest at nightfall or nighttime for Onc. Rosy Sunset. It is worth noting that (E)-4,8-dimethyl-l,3,7-nonatriene, which is generally emitted from the damaged leaves and attracts insects, is only found in a few flowers, such as Magnolia taxa and Liriodendron tulipifera [26]. In our results, (E)-4,8-dimethyl-l,3,7-nonatriene was emitted in highest concentrations at nighttime, although it is not known whether this is related to the attraction of nocturnal insects. Linalool was a primary component of emission, over all times of the daily rhythm.
In terms of horticultural characteristics, the finding that the Onc. Rosy Sunset odor was strongest in the morning can be used to market the flowers to consumers and as a reference for hybrid offspring selection time. In addition, the selection of parent plants that emit aroma during the afternoon or nightfall can be used in breeding programs with Onc. Rosy Sunset to breed new Oncidium cultivars that emit aroma all day.

Changes of Volatile Components with the Flowering Period
The quantity of volatile components increased with the flowering period, peaking during stages 3-6 of the inflorescence and decreasing after stage 7, with almost no odor at stage 9 ( Figure 2). The level of emitted volatile components changes with the flowering period, with the strongest odor during stages 3-6 in Onc. Rosy Sunset and not in the initial flowering stage. This result was repeated in a variety of flowers. For comparison, the flowering period of Phal. Nobby's Pacific Sunset was 51 days, and the odor was strongest at 8-20 days after flowering (DAF) [8]. Phal. bellina emitted the highest levels of geraniol and linalool at 5-7 DAF, and the flowering period was 14 days [6]. The odor was strongest at 2 or 3 DAF for Narcissus tazetta var. chinensis with a 6-day flowering period [17].
Four of the major volatile compounds were linalool, benzaldehyde, β-myrcene, and β-caryophyllene, and their relative percentage change in composition with the flowering period is shown in Figure 3. The relative percentage of linalool was much higher than that of other volatile components during all phases of the Onc. Rosy Sunset flowering period, and the level significantly increased during the first flower bloom of the inflorescence. These changes were similar to those in the total relative content. The relative percentage of benzaldehyde was higher during stages 7-8, which may be affected by the decrease in the relative quantity of linalool. No significant changes of β-myrcene and β-caryophyllene were observed during the flowering period ( Figure 3). Onc. Sharry Baby had a higher quantity of volatile compounds during the flowering stage than in the initial flowering stage, and the relative content of alkenes, alcohol, and esters was found to increase with the flowering stage [10]. However, the total relative content significantly lowers at stage 5 in our experiment, but rises again at stage 6, presumably due to the colder weather. Although the experiment temperature was controlled, the environment still varied in temperature because of a cold snap. Zhang et al. [12] reported that the temperature will affect the volatile emissions in Onc. Sharry Baby, and the amount of aroma components and total relative content were both higher at 30 • C than at 10 • C. Adachi et al. [22] also indicated that for Gongora bufonia, the odor was strongest at 10:00, which was the period in which the temperature began to rise.

Volatile Components in Different Parts of Onc. Rosy Sunset Flowers
The Onc. Rosy Sunset flowers were divided into petal, sepal, lip, and column parts to analyze the differences in their volatile emission profile compositions. The results showed that the total relative content and amounts of volatile compounds in the petals and sepals were significantly higher than in the lip and column. A total of 43 compounds were identified from the petals and sepals, with a total relative content count of 12,798 and 11,120, respectively. A total of 27 and 25 compounds were found from the lip and column regions, respectively, with the total relative content count at 859 and 716, respectively (Table 3). Phal. bellina emitted odor from the perianth and not the column [5]. In flowering plants, scent glands are called osmophores. In the orchid, odors are not produced by all flower parts, however, and the osmophores are found on the sepals, petals, or parts of the lip, mainly at the base of the lip [27,28], as seen in Vanilla edwallii and Gongora bufonia [22,29]. The most significant structure of Oncidium flowers is the lip, and the callus at the base of the lip is often considered the location of the osmophores. Zhang et al. [10] reported that the lip of Onc. Sharry Baby was the location where the majority of the odor was emitted.  Although hydrocarbons were the most common type of volatile compounds from the four flower parts, alcohols had the highest total relative content in the petals and sepals ( Table 3). As shown in Figure 4, the major component found in the petals and sepals was linalool (classed as an alcohol), which was shown to be the strongest odor in two parts. However, the strongest odor was detected from the lip of Onc. Sharry Baby, and the column had the most complex volatile compounds [10]. The major volatile compounds were also found to differ between the four parts, with linalool found as the major compound in the petals and sepals; benzaldehyde and β-myrcene had the highest relative percentage in the lip, while β-caryophyllene was the highest in the column ( Figure 4). Meanwhile, Phal. Nobby's Pacific Sunset was reported to have the highest total relative content of volatile compounds in the petals and dorsal sepals. The major volatile compound in all five flower parts studied was linalool [8]. There is, however, a different distribution of major volatile compounds in different flower parts of Onc. Rosy Sunset. It is thought that the location of the osmophores and structure changes between the orchid species and different cultivars, even in Oncidium, creates a large effect.  Although hydrocarbons were the most common type of volatile compounds from the four flower parts, alcohols had the highest total relative content in the petals and sepals ( Table 3). As shown in Figure 4, the major component found in the petals and sepals was linalool (classed as an alcohol), which was shown to be the strongest odor in two parts. However, the strongest odor was detected from the lip of Onc. Sharry Baby, and the column had the most complex volatile compounds [10]. The major volatile compounds were also found to differ between the four parts, with linalool found as the major compound in the petals and sepals; benzaldehyde and β-myrcene had the highest relative percentage in the lip, while β-caryophyllene was the highest in the column ( Figure 4). Meanwhile, Phal. Nobby's Pacific Sunset was reported to have the highest total relative content of volatile compounds in the petals and dorsal sepals. The major volatile compound in all five flower parts studied was linalool [8]. There is, however, a different distribution of major volatile compounds in different flower parts of Onc. Rosy Sunset. It is thought that the location of the osmophores and structure changes between the orchid species and different cultivars, even in Oncidium, creates a large effect. Although hydrocarbons were the most common type of volatile compounds from the four flower parts, alcohols had the highest total relative content in the petals and sepals ( Table 3). As shown in Figure 4, the major component found in the petals and sepals was linalool (classed as an alcohol), which was shown to be the strongest odor in two parts. However, the strongest odor was detected from the lip of Onc. Sharry Baby, and the column had the most complex volatile compounds [10]. The major volatile compounds were also found to differ between the four parts, with linalool found as the major compound in the petals and sepals; benzaldehyde and β-myrcene had the highest relative percentage in the lip, while β-caryophyllene was the highest in the column ( Figure 4). Meanwhile, Phal. Nobby's Pacific Sunset was reported to have the highest total relative content of volatile compounds in the petals and dorsal sepals. The major volatile compound in all five flower parts studied was linalool [8]. There is, however, a different distribution of major volatile compounds in different flower parts of Onc. Rosy Sunset. It is thought that the location of the osmophores and structure changes between the orchid species and different cultivars, even in Oncidium, creates a large effect.  Figure 1; y RI: literature retention indices references were checked for all compounds on DB-1 column; x RI: retention indices obtained using series of n-alkanes (C5-C25) on DB-1 column; w data are mean ± SD of three replicates.   In this study, the results showed that daily rhythm, flowering period, and temperature affected aroma emissions. These results were also showed in Onc. Sharry Baby; the odor was strongest during the flowering stage and in a 30 °C environment, on the other hand, the lower temperature reduced the number of volatile compounds and relative content [10,12]. The qualitative and quantitative differences between the present study and those from other parts of the world may be attributable to the differences in various extraction techniques, cultivated varieties, and ecological and climatic conditions [30]. Therefore, we set the conditions at 25 °C, 5-8 DAF, and extraction was performed during the 10:00-12:00 AM period for the more stable analytical performance.

Plant Materials
Onc. Rosy Sunset is an ornamental potted Oncidium with a sweet scent, which was purchased from Yung Hsin Orchids, Taichung, Taiwan. Volatile components were identified during the flowering period, cultivated in the National Museum of Natural Science cold room greenhouse, controlled at 25 ± 3 °C.

Volatile Components of Onc. Rosy Sunset
The volatile compounds extracted from the Onc. Rosy Sunset flowers were analyzed by headspace solid-phase microextraction (HS-SPME) coupled with gas chromatography (GC) and gas chromatography-mass spectrometry (GC-MS). In a study using five different SPME fibers, volatile compounds were analyzed in Phal. Nobby's Pacific Sunset. The results show that fibers with 50/30 μm divinylbenzene/carboxen/polydimethylsiloxane (DVB/CAR/PDMS) had optimal extraction ability, with linalool as one of the major volatile compounds [8]. This study also used 50/30 μm DVB/CAR/PDMS fibers for better analytical performance.
HS-SPME of volatile components in Onc. Rosy Sunset was performed using fiber adsorption techniques, and subsequent analysis was carried out using GC (gas chromatography, 6890 GC, Agilent, Santa Clara, CA, USA) and GC-MS (gas chromatography-mass spectrometry, 6890 GC-5973N MSD, Agilent, Santa Clara, CA, USA). At 10:00-12:00 AM Onc. Rosy Sunset inflorescence with 10 flowers, 5-8 DAF, was enclosed with a glass vessel (9 cm × 6.5 cm) and sealed with paraffin film, the device is shown in Figure 5. SPME fiber (50/30 μm DVB/CAR/PDMS, Supelco, Bellefonte, PA, USA) was exposed to the inflorescence for 40 min in order to extract the aroma components in a 25 ± 3 °C air In this study, the results showed that daily rhythm, flowering period, and temperature affected aroma emissions. These results were also showed in Onc. Sharry Baby; the odor was strongest during the flowering stage and in a 30 • C environment, on the other hand, the lower temperature reduced the number of volatile compounds and relative content [10,12]. The qualitative and quantitative differences between the present study and those from other parts of the world may be attributable to the differences in various extraction techniques, cultivated varieties, and ecological and climatic conditions [30]. Therefore, we set the conditions at 25 • C, 5-8 DAF, and extraction was performed during the 10:00-12:00 AM period for the more stable analytical performance.

Plant Materials
Onc. Rosy Sunset is an ornamental potted Oncidium with a sweet scent, which was purchased from Yung Hsin Orchids, Taichung, Taiwan. Volatile components were identified during the flowering period, cultivated in the National Museum of Natural Science cold room greenhouse, controlled at 25 ± 3 • C.

Volatile Components of Onc. Rosy Sunset
The volatile compounds extracted from the Onc. Rosy Sunset flowers were analyzed by headspace solid-phase microextraction (HS-SPME) coupled with gas chromatography (GC) and gas chromatography-mass spectrometry (GC-MS). In a study using five different SPME fibers, volatile compounds were analyzed in Phal. Nobby's Pacific Sunset. The results show that fibers with 50/30 µm divinylbenzene/carboxen/polydimethylsiloxane (DVB/CAR/PDMS) had optimal extraction ability, with linalool as one of the major volatile compounds [8]. This study also used 50/30 µm DVB/CAR/PDMS fibers for better analytical performance.
HS-SPME of volatile components in Onc. Rosy Sunset was performed using fiber adsorption techniques, and subsequent analysis was carried out using GC (gas chromatography, 6890 GC, Agilent, Santa Clara, CA, USA) and GC-MS (gas chromatography-mass spectrometry, 6890 GC-5973N MSD, Agilent, Santa Clara, CA, USA). At 10:00-12:00 AM Onc. Rosy Sunset inflorescence with 10 flowers, 5-8 DAF, was enclosed with a glass vessel (9 cm × 6.5 cm) and sealed with paraffin film, the device is shown in Figure 5. SPME fiber (50/30 µm DVB/CAR/PDMS, Supelco, Bellefonte, PA, USA) was exposed to the inflorescence for 40 min in order to extract the aroma components in a 25 ± 3 • C air conditioned laboratory, after which each sample was injected into a gas chromatograph injection unit. This experiment and all other experiments in this study were carried out in triplicate.
Molecules 2017, 22, 1468 10 of 12 conditioned laboratory, after which each sample was injected into a gas chromatograph injection unit. This experiment and all other experiments in this study were carried out in triplicate.

Changes of Volatile Components with Daily Rhythm
The Onc. Rosy Sunset inflorescence with 10 flowers, 5-8 DAF, was collected and analyzed for volatile components at two hour intervals between 10:00 and 08:00 (inclusive). The SPME method, as described above, was carried out for each period in triplicate, and each sample monitored the changes for 24 h. During non-adsorption times, the paraffin film was opened to ventilate the glass vessel.

Volatile Component Changes during the Flowering Period
For the flower buds of the Onc. Rosy Sunset, 10 flower inflorescences were selected and defined as follows: stage 0, flower buds turned color; stage 1, 50% of flowers in bloom; stage 2, full inflorescence flowering; stage 3-8, every 2 days that have elapsed post flowering; stage 9, the first withered flower observed; and stage 10, 50% of flowers withered. The SPME method, as described above, was performed during the 10:00-12:00 AM period.

Volatile Components in Different Parts of the Onc. Rosy Sunset Flowers
The 5-8 DAF flowers were picked and separated into four parts: petal, sepal, lip, and column. The four parts of ten flowers were separated into glass bottles (precleaned # 27343 22-mL clear screw cap vials; Supelco Inc.). The SPME method was used to adsorb volatiles for 60 min.

GC and GC-MS
A GC equipped with a 60 m × 0.25 mm i.d column was used to perform qualitative and quantitative analysis of the volatile compounds. The column used a DB-1 fused-silica capillary with a film thickness of 0.25 μm and a flame ionization detector. The injector and detector temperatures were maintained at 250 and 300 °C, respectively. The sample was held at 40 °C for 1 min and raised to 200 °C at 2 °C/min then held for 9 min at the temperature of oven. The nitrogen carrier gas flow rate was 1 mL/min. Kovats indices were calculated for the separated components relative to a C5-C25 n-alkanes mixture [31]. The peak area normalization measurements were used to calculate the percentage composition.

Changes of Volatile Components with Daily Rhythm
The Onc. Rosy Sunset inflorescence with 10 flowers, 5-8 DAF, was collected and analyzed for volatile components at two hour intervals between 10:00 and 08:00 (inclusive). The SPME method, as described above, was carried out for each period in triplicate, and each sample monitored the changes for 24 h. During non-adsorption times, the paraffin film was opened to ventilate the glass vessel.

Volatile Component Changes during the Flowering Period
For the flower buds of the Onc. Rosy Sunset, 10 flower inflorescences were selected and defined as follows: stage 0, flower buds turned color; stage 1, 50% of flowers in bloom; stage 2, full inflorescence flowering; stage 3-8, every 2 days that have elapsed post flowering; stage 9, the first withered flower observed; and stage 10, 50% of flowers withered. The SPME method, as described above, was performed during the 10:00-12:00 AM period.

Volatile Components in Different Parts of the Onc. Rosy Sunset Flowers
The 5-8 DAF flowers were picked and separated into four parts: petal, sepal, lip, and column. The four parts of ten flowers were separated into glass bottles (precleaned # 27343 22-mL clear screw cap vials; Supelco Inc.). The SPME method was used to adsorb volatiles for 60 min.

GC and GC-MS
A GC equipped with a 60 m × 0.25 mm i.d column was used to perform qualitative and quantitative analysis of the volatile compounds. The column used a DB-1 fused-silica capillary with a film thickness of 0.25 µm and a flame ionization detector. The injector and detector temperatures were maintained at 250 and 300 • C, respectively. The sample was held at 40 • C for 1 min and raised to 200 • C at 2 • C/min then held for 9 min at the temperature of oven. The nitrogen carrier gas flow rate was 1 mL/min. Kovats indices were calculated for the separated components relative to a C5-C25 n-alkanes mixture [31]. The peak area normalization measurements were used to calculate the percentage composition.
The volatile compounds were identified using a GC-MS. The GC was equipped as described above and a DB-1 fused-silica capillary column with a film thickness of 0.25 µm coupled to an MS. The injector temperature was held at 250 • C. The helium carrier gas flow rate was 1 mL/min. The ionization energy was 70 eV at 230 • C. A mass spectrometry (MS) library (Wiley 7N) was used to identify the constituents. Comparison of the GC and GC-MS analysis results was completed to identify compounds qualitatively and quantitatively. In addition, the constituents were confirmed by comparing the Kovats indices or GC retention time data with those of authentic standards or published in the literature.

Statistical Analysis
The ANOVA mean value among treatments was compared with the LSD multiple range at a 5% level of significance. This was performed using CoStat6.1 software (CoHort software, Minneapolis, Mn, USA).

Conclusions
The major aroma volatile compounds of Onc. Rosy Sunset were linalool, benzaldehyde, β-myrcene, tiglaldehyde, benzyl alcohol, and nerolidol. The odor was strongest around noon three days after full flowering, and the main source of aroma was from the petals and sepals. This paper is the first comprehensive study of the change in volatile aroma compounds with daily rhythm, flowering period, and different flower parts for Oncidium. In the application of horticulture, this is useful information for commodity sales, understanding the breeding of parental characteristics such as aroma composition, and as a reference for future research on the location and structural characteristics of osmophores in Oncidium.