Journal Search Engine
Search Advanced Search Adode Reader(link)
Download PDF Export Citaion korean bibliography PMC previewer
ISSN : 1225-5009(Print)
ISSN : 2287-772X(Online)
Flower Research Journal Vol.29 No.3 pp.138-145

Identification of the Change in Volatile Compound Composition according to the Flowering Stages of Wild Chrysanthemum Species Using HS-SPME-GC-MS

Seung Won Kang*
Faculty of Life and Environmental Sciences, University of Tsukuba, Tsukuba, 305-8577, Japan
* Corresponding author: Seung won Kang Tel: +81-29-853-4807 E-mail:
28/07/2021 01/08/2021 04/08/2021


The change in volatile compound composition of three wild Chrysanthemum species (Chrysanthemum boreale, C. indicum, and C. indicum var. albescens) was identified and analyzed according to four flowering stages using HS-SPME-GC-MS (headspace solid-phase microextraction coupled to gas chromatography–mass spectrometry). The top five compounds of each flowering stage were selected because those main compounds accounted for 43.25%, 44.14%, and 54.20% of the total relative content of volatile compounds from C. boreale, C. indicum, and C. indicum var. albescens, respectively. Nine compounds (1S-α-pinene, α-thujone, chrysanthenone, umbellulone, thymol, caryophyllene, germacrene D, α-zingiberene, and α-patchoulene) in C. boreale were ranked in the top five compounds through the whole flowering stages. In C. indicum, camphene, eucalyptol, camphor, umbellulone, bornyl acetate, caryophyllene, β-farnesene, germacrene D, and α-zingiberene were ranked in the top five compounds. However, only five compounds (camphor, bornyl acetate, β-farnesene, germacrene D, and α-zingiberene) were ranked in C. indicum var. albescens showing a more stable composition rather than C. boreale and C. indicum. Flowerheads of three wild Chrysanthemums showed a different profile of volatile compounds according to different flowering stages, varying compositions, and relative content in the top five volatile compounds. This study illustrates how main volatile compounds in wild Chrysanthemums change dynamically during the flowering regarding compositions and their relative contents, suggesting that it should provide a useful index for harvesting or blending certain target compounds from wild Chrysanthemums.



    Asteraceae (Compositae) is the one of the largest plant family in flowering plants with more than 23,000 species. The genera Chrysanthemum has been widely used for cut flowers and potted plants in floricultural industry over the world. Some wild Chrysanthemums, such as Chrysanthemum. boreale, C. indicum, and C. indicum var. albescens has been widely used as medicinal plants in Korea and China (Shao et al. 2020). Flavonoids extracted from C. boreale, such as acacetin, apigenin, and luteolin, were known to inhibit rat lens aldose reductase (RLAR) activity (Shin et al. 1995).

    In recent, SPME (solid phase micro extraction) method are increasingly used to analyze volatile compounds because SPME is known as organic solvent free technique and also independent to sample type so it can extract volatile compounds from solid or aqueous samples (Arthur and Pawliszyn 1990;Lord and Pawliszyn 2000). In addition, SPME can extract with very high sensitivity to volatile compounds ranging from nanogram to microgram and whole extracts can be injected to GC without leakage and is easy to handle from extraction to analysis of volatile compounds (Dong et al. 2007). Furthermore, SPME enables routine analysis using autosampler when it is coupled to GC or GC-MS providing high accuracy and reproducibility.

    In addition, the essential oil of C. boreale contains camphor, α-thujone, β-caryophyllene, 1,8-cineole, α-pinene, piperitone, β-pinene, camphene, and cis-chrysanthenol, and is known to exhibit antibacterial activity and suppressed activity of Streptococcus mutans by inhibiting virulence factors (Kim et al. 2003;Kim et al. 2015). Essential oil of C. indicum also contains α-pinene, 1,8-cineol, α-thujene, camphor, etc., but it contains different volatile compounds, such as terpinen-4-ol, bornyl acetate, borneol, germacrene D, and α-cadinol (Chang and Kim 2008;Jung 2009). C. indicum var. albescens is a variety of C. indicum and has white flowers rather than yellow color that C. boreale and C. indicum have, but shows very similar phenotypic characteristics to both wild Chrysanthemums. However, unlike C. boreale and C. indicum, there are a few reports on volatile compounds of C. indicum var. albescens. Leaf of C. indicum var. albescens contains borneol, phytol, germacrene D, and α-pinene, and camphene (Kim et al. 2014). However, despite volatile compounds of wild Chrysanthemums and their biological functions have been well elucidated, few researches attempted to improve the volatile compounds of wild Chrysanthemums. Therefore, to extract important or target compounds, it is necessary to understand flowering stage specific composition of volatile compounds both in quality and in quantity

    This study aims to identify change of volatile compounds in different flowering stages. Therefore, in this study, volatile compounds produced from three wild Chrysanthemums, C. boreale, C. indicum, and C. indicum var. albescens were investigated not only to identify volatile compounds but also to analyze change of volatile compounds according to flowering stages using HS-SPME-GC-MS.

    Materials and Methods

    Plant material and extraction of volatile compounds

    Seeds of Chrysanthemum boreale (C. seticuspe), C. indicum, and the variety of C. indicum (var. albescens) were obtained from the farm Sansanguideulgukhwahyanggi (Gangwon, Republic of Korea) and cultivated in the experimental field at Chung-Ang University, Anseong, Republic of Korea (37°00'02.4"N 127°13'41.1"E) from April to November 2011. Plants were grown in natural condition without any additional fertilization to keep as close to natural environment. Flowerheads of three wild Chrysanthemums were harvested at the end of October 2011 according to the four flowering stages; Before Flowering (BF), Begin to open (BO), Half open (HO), and Fully open (Fig. 1). Ten to twenty flowerheads of BF, BO, HO, and FO stages, respectively, were pooled and immediately immersed into liquid nitrogen and freeze-dried at -50°C for 72 h, ground using Automill (Tokken, Inc., Japan) at 1,000 RPM for 1 min, and finally stored in a 20 mL of headspace screw top vial (C4020-18, Thermo Scientific) with desiccants at -80℃ until use.

    Extraction of volatile compounds

    To extract volatile compounds, sample preparation procedure was followed as Tikunov et al. (2005) with slight modification. 50 mg of lyophilized powder were added to a 20 mL headspace screw vial. 500 μL of 100 mM EDTA-NaOH solution was added and then 1.5 g CaCl2・ 2H2O were added immediately. Finally, the sample vial was sonicated at room temperature for 5 min and then placed to the Multi-Purpose Sampler (MPS 2L-XT, GERSTEL, Germany) equipped to GC-MSD (GC-7890A, MSD-5975C) for analysis.

    Measurement of volatile compounds of using HS-SPME -GC-MSD

    Samples were prepared from three wild Chrysanthemum species and four flowering stages, respectively. Sample vials were agitated at 65°C for 20 min to saturate sample vials with volatile compounds and then volatile compounds were absorbed using SPME-fiber (solid phase micro extraction, PDMS/DVB) for 20 min. SPME fiber was thermally cleaned twice (pre- and post-injection) for 2 min at 250°C in GC injection port and injected into GC splitless injection port.

    GC-MS analysis was performed using Agilent GC 7890A gas chromatograph equipped with a HP-5MS (30 m × 0.25 mm × 0.25 μm film thickness, Agilent, USA) capillary column. Oven temperature was programmed initially at 40°C for 1 min then raised to 250°C at a rate of 5°C·min-1. Injection temperature was 250°C. Interface line temperature of mass spectrometer 5975C (Agilent, USA) coupled with GC was programmed at 280°C.

    Data analysis

    Mass spectra were recorded under 70eV (EI) and identified by comparison of m/z data and retention index with NIST05 Mass spectral library database (NIST, USA) and NIST Chemistry WebBook ( Quantity of selected volatile compounds was analyzed by area normalization method using MSD Chemstation (Version E.02.02.1431, Agilent Technologies, Inc.). Retention index were calculated based on the equation suggested by van Den Dool and Kratz (1963) and gas chromatographic retention data ( using Alkane standard solution C8-C20 (04070, Sigma-Aldrich, USA).

    To evaluate differences in the composition and relative content of volatile compounds of three wild Chrysanthemums through four flowering stages, One-way analysis of variance (ANOVA) was performed with Tukey’s HSD as post-hoc analysis at p < 0.05 using IBM SPSS Statistics version 25 (n = 3). In addition, OPLS-DA and HCA were performed using MetaboAnalyst 5.0 (; Pang et al. 2021) for selected 14 main volatile compounds that are ranked in the top five to identify volatile compounds contributed to differences among four flowering stages of three wild Chrysanthemums. Euclidean distance and Ward`s linkage method were used to draw dendrogram (Xia et al. 2009).

    Results and Discussion

    About 130 peaks were detected from the flowers of three wild Chrysanthemum species, C. boreale, C. indicum, and c. indicum var. albescens and 100 out of 130 volatile compounds were identified using NIST library with over 90% of matching identity. Relative content of the top five volatile compounds of C. boreale, C. indicum, and C. indicum var. albescens accounted for 43.25, 44.14, and 54.20% respectively throughout all four flowering stages (Fig. 2). In addition, top 30 volatile compounds of C. boreale occupied 82.04% of total volatile compounds flowers. The top 30 volatile compounds of C. indicum and C. indicum var. albescens also occupied 81.02 and 85.00% respectively. These results suggest small number of volatile compounds dominates large portion of volatile compounds. Therefore, in this study, the top five volatile compounds found in different flowering stages from three wild Chrysanthemums were selected and analyzed.

    Volatile compound change of wild Chrysanthemums according to flowering stages

    Throughout the whole flowering stages of C. c, the nine volatile compounds such as; umbellulone (3-thujen-2-one), chrysanthenone, thymol, α-patchoulene, α-zingiberene, caryophyllene, germacrene D, α-thujone, and 1S-α-pinene were ranked in the top five volatile compounds (Table 1). It was found that relative content of volatile compounds emitted from flowers changed according to flowering stages and also their rank changed except for the top two volatile compounds, umbellulone and chrysanthenone. The major volatile compound was umbellulone ranging from with 23.63 to 27.84% showing slight decrease through all four flowering stages. The second was chrysanthenone and accounted for 6.10 to 7.19%. The other three main compounds found at the BF stage were α-thujone (5.74%), α-patchoulene (3.42%), and germacrene D (3.40%). Peak area of α-thujone, the third main compound at BF stage, steeply decreased from HO stage presenting high fluctuation to FO stage with 0.58, 2.49, and 0.71%, respectively resulting in exclusion from top five main compounds. Peak area of α-patchoulene ranged from 3.42 to 4.14% and remained as top five volatile compounds like umbellulone and chrysanthenone. Germacrene D was the fifth compound with 3.40% but it slightly decreased to 2.41 to 2.67% and excluded from the top five compounds except for BF stage.

    In C. indicum, top five volatile compounds included nine compounds (camphor, umbellulone, eucalyptol, germacrene D, β-farnesene, caryophyllene, α-zingiberene, bornyl acetate, and camphene) (Table 2). At the BF stage, the top five compounds were camphor (26.62%), camphene (6.05%), eucalyptol (5.93%), caryophyllene (5.71%), and germacrene D (4.55%). Camphor accounted for 26.62% at the BF stage but relative content gradually decreased to 16.83% at the FO stage while holding the top rank during the all flowering stages. On the other hand, the other top four compounds drastically changed through four flowering stages. Camphene, the top second volatile compound, accounted for 6.05% but the amount steeply decreased from BO to FO stage ranging 2.04 to 3.53% resulting in exclusion from top five volatile compounds except for BF stage. Relative amount of eucalyptol decreased from 5.93 to 3.97% from BF to HO stage, but it recovered to the similar content by increasing to 5.00% at FO stage. The amount of caryophyllene decreased from 5.71 to 3.50% between BF and BO stage, but then increased to 4.23 and 4.32% between HO and FO stage, respectively. Germacrene D showed similar tendency like caryophyllene by decreasing its amount from 4.55 to 3.76% but increased to 5.20 and 4.94% between HO and FO stage.

    The top five volatile compounds of C. indicum var. albescens included camphor, bornyl acetate, β-Farnesene, α -Zingiberene, and Germacrene D (Table 3). Unlike C. boreale and C. indicum, the top five volatile compounds did not change through all four flowering stages, but rank among the other four volatile compounds changed except for camphor. The top first volatile compound was camphor (22.68 to 26.67%) like C. indicum but it maintained the similar amount during the flowering. The second compound was Bornyl acetate with 11.55 and 11.09% between BF and BO stages, but its amount steeply decreased from HO to 4.99 % resulting in dropping to fifth from HO and FO stages. On the other hand, amount of β-Farnesene ranked to the fifth with 5.38 and 5.53% at the beginning of flowering but it steeply increased to 11.17 and 11.02% as the flowering progressed. The other two top five volatile compounds were germacrene D and α-zingiberene. Germacrene D tended to decrease while α-zingiberene increased as flowering progressed resulting in reversion of rank between the third to fourth.

    Differentiation of volatile compounds according to flowering stage is known to reduce use of resources after pollination (Dudareva et al. 1996) resulting in dynamic change of floral scents emitted from Protea species, Luculia pinceana, and Rosa hybrida (Lee et al. 2008;Li et al. 2016;Steenhuisen et al. 2010). In addition, application of soil amendments containing high amount of calcium or calcium itself is known to increase terpenoid contents in flower head and essential oil of C. boreale (Lee et al. 2005;Lee and Yang 2006).

    Identification of volatile compounds contributed to difference of composition

    Considering nine main volatile compounds ranked in the top five in C. boreale and C. indicum, respectively, and five compounds of C. indicum var. albescens, it was found that total 14 volatile compounds were ranked in top five throughout the whole flowering stages (Fig. 3). Only two out of 14 compounds, germacrene D and α-zingiberene, were ranked in the top five compounds from all three Chrysanthemums. Umbellulone and caryophyllene were found only between C. boreale and C. indicum while camphor, bornyl acetate, and β-farnesene were ranked only between C. indicum and C. indicum var. albescens. On the other hand, no compounds were found in top five between C. boreale and C. indicum var. albescens. As a result, the pattern of changing relative content of volatile compounds and composition was found to be similar between C. boreale and C. indicum from HCA using 14 volatile compounds, but C. indicum var. albescens is a little bit further from C. boreale and C. indicum (Fig. 4).

    To identify contribution of volatile compounds that affected the differences among three wild Chrysanthemums according to four different flowering stages, OPLS-DA was used as multivariate analysis and volatile compounds whose VIP-score is larger than 1.0 and P-value is lower than 0.05 were extracted (Fig. 5). In result, six volatile compounds, camphor (1.249), umbellulone (1.245), chrysanthenone (1.226), bornyl acetate (1.149), β-Farnesene (1.133), and camphene (1.097), were extracted. Umbellulone and chrysanthenone are the main volatile compounds emitted from the flowers of C. boreale, camphor, bornyl acetate, and β-farnesene were the main volatile compounds of both C. indicum and C. indicum var. albescens.

    It was found that both in the quantity and quality of volatile compounds drastically change as flowering progressed in three wild Chrysanthemum species. The current study will provide the optimum flowering stage to harvest flowerheads producing certain target compounds and also will be the first report on the dynamic change of volatile compounds according to flowering stage of flowerheads from three wild Chrysanthemums. Therefore, the quality and quantity of extracts or essential oils could be improved by selecting a certain flowering stage or combining flowerhead of different flowering stages.


    This work was supported by the National Research Foundation of Korea Grant funded by the Korean Government (Ministry of Education, Science and Technology). [NRF- 2010-355-F00009]



    Four flowering stages of chrysanthemum species used in this experiment. Photos were obtained using flowerheads of C. indicum. The bars in figures represents 5 mm.


    Relative contents of volatile compounds ranked in the top 5, 6 to 10, 11 to 20, 21 to 30, and others from three wild Chrysanthemums. A: C. boreale, B: C. indicum, C: C. indicum var. albescens.


    The top five volatiles overlapped among three wild Chrysanthemum species through four different flowering stages. Four out of nine compounds of C. boreale were overlapped with C. indicum and seven out nine volatile compounds of C. indicum were overlapped with C. boreale and C. indicum var. albescens.


    Dendrogram obtained after hierarchical cluster analysis of the volatile compounds of three wild Chrysanthemums. The distances between samples were calculated by Euclidean distance and Ward`s linkage method was used. The letters of Cb, Ci, and CiA represent C. boreale, C. indicum, and C. indicum var. albescens, respectively and two numbers after letters represent sample numbers. BF: before flowering, BO: begin to open, HO: half open, FO: fully open.


    Distribution of variable importance for prediction (VIP) scores of the top ten main volatile compounds with p < 0.05 identified by OPLS-DA from three wild Chrysanthemums.


    The nine Volatiles of C. boreale ranked in the top five main compounds according to flowering stages.

    The nine volatiles of C. indicum ranked in the top five main compounds according to flowering stages.

    The five volatiles of C. indicum var. albescens ranked in the top five main compounds according to flowering stages.


    1. Arthur CL , Pawliszyn J. (1990) Solid phase microextraction with thermal desorption using fused silica optical fibers. Anal Chem 62:2145-21488
    2. Chang KM , Kim GH (2008) Volatile aroma composition of Chrysanthemum indicum L. flower oil. J Food Sci Nut 13:122-127
    3. Dong L , Wang J , Deng C , Shen X (2007) Gas chromatographymass spectrometry following pressurized hot water extraction and solid-phase microextraction for quantification of eucalyptol, camphor, and borneol in Chrysanthemum flowers. J Sep Sci 30:86-89
    4. Dudareva N , Cseke L , Blanc VM , Pichersky E (1996) Evolution of floral scent in Clarkia : Novel patterns of S-linalool synthase gene expression in the C. breweri flower. The Plant Cell 8:1137–1148
    5. Jung EK (2009) Chemical composition and antimicrobial activity of the essential oil of Chrysanthemum indicum against oral bacteria. J Bacteriol Virol 39:61-69
    6. Kim BS , Park SJ , Kim MK , Kim YH , Lee SB , Lee KH , Choi NY , Lee YR , Lee YE , You YO (2015) Inhibitory effects of Chrysanthemum boreale essential oil on biofilm formation and virulence factor expression of Streptococcus mutans. Evidence-based Comple Alt Med Article ID 616309
    7. Kim KJ , Kim YH , Yu HH , Jeong SI , Cha JD , Kil BS , You YO (2003) Antibacterial activity and chemical composition of essential oil of Chrysanthemum boreale. Planta Med 69:274-277
    8. Kim SJ , Ha TJ , Kim JY , Nam JH , Yoo DL , Suh JT , Kim KS (2014) Volatile flavor compounds in the leaves of fifteen taxa of Korean native Chrysanthemum species. Korean J Hort Sci Biotechnol 32:558-570
    9. Lee KD , Yang MS (2006) Soil amendment effects on the yield and terpene contents of the flowerhead of Chrysanthemum boreale M. Agrochimica 50:62-71
    10. Lee KD , Yang MS , Supanjania Smith, DL (2005) Fertilizer effect on the yield and terpene components from the flowerheads of Chrysanthemum boreale M. (Compositae). Agron Sustain Dev 25:205-211
    11. Lee YS , Kim SJ , Kim KJ (2008) Analysis of volatile compounds according to the flowering stages of rose using gas-chromatography mass spectrometry. Korean J Hort Sci Technol 26:501-507
    12. Li Y , Ma H , Wan Y , Li T , Liu X , Sun Z , Li Z (2016) Volatile organic compounds emissions from Luculia pinceana flower and its changes at different stages of flower development. Molecules 21:531
    13. Lord H , Pawliszyn J (2000) Evolution of solid-phase microextraction technology - A review. J Chromatogr A 885:153-193
    14. Pang Z , Chong J , Zhou G , Morais DA de L , Chang L , Barrette M , Gauthier C , Jacques PÉ , Li S , Xia J (2021) MetaboAnalyst 5.0: Narrowing the gap between raw spectra and functional insights. Nuc Acids Res 49: W388–W396
    15. Shao Y , Sun Y , Li D , Chen Y (2020) Chrysanthemum indicum L.: A comprehensive review of its botany, phytochemistry and pharmacology. Amer J Chinese Med 48:1-27
    16. Shin KH , Kang SS , Seo EA , Shin SW (1995) Isolation of aldose reductase inhibitors from the flowers of Chrysanthemum boreale. Arch Phar Res 18:65-68
    17. Steenhuisen SL , Raguso RA , Jürgens A , Johnson SD (2010) Variation in scent emission among floral parts and inflorescence developmental stages in beetle-pollinated Protea species (Proteaceae). South Afr J Bot 76:779-787
    18. Tikunov Y , Lommen A , Ric de Vos CH , Verhoeven HA , Bino RJ , Hall RD , Bovy AG (2005) A novel approach for nontargeted data analysis for metabolomics. Large-scale profiling of tomato fruit volatiles. Plant Physiol 139: 1125–1137
    19. van Den Dool H , Kratz DP (1963) A generalization of the retention index system including linear temperature programmed gas-liquid partition chromatography. J Chromatography A 11:463-471
    20. Xia J , Psychogios N , Young N , Wishart DS (2009) MetaboAnalyst: A web server for metabolomic data analysis and interpretation. Nuc Acids Res 37:W652–W660