Biochemical Systematics and Ecology 29 (2001) 267–285
Variation in volatile compounds from tansy
(
Tanacetum vulgare
L.) related to genetic and
morphological differences of genotypes
Marjo Keskitalo
a,b,*, Eija Pehu
a, James E. Simon
ba
Department of Plant Production, P.O. Box 27, University of Helsinki, FIN-00014, Finland
b
Department of Horticulture, Center for New Crops and Plant Products, Purdue University, West Lafayette, IN 47906-1165, USA
Received 18 January 1999; accepted 15 February 2000
Abstract
Air-dried flower heads of 20 Finnish tansy genotypes were extracted with petroleum ether and analyzed using GC–MS. A total of 55 volatile compounds were detected, and 53 were identified. Of the tansy genotypes studied, 15 were well defined and five were mixed chemotypes. Complete linkage analysis differentiated the populations into six clusters. The most frequently found monoterpene was camphor with or without several satellite compounds such as camphene, 1,8-cineole, pinocamphone, chrysanthenyl acetate, bornyl acetate and isobornyl acetate. In 13 genotypes, camphor concentration exceeded 18.5% and in seven genotypes, camphor was less than 7.2%. Other chemotypes rich intrans thujone, artemisia ketone, 1,8-cineole, or davadone-D were also identified. Davadone-D and a mixed chemotype, containing tricyclene and myrcene, were identified from a Finnish tansy for the first time. Geographically, most chemotypes containing camphor originated from Central Finland, whereas chemotypes without camphor such as artemisia ketone, davadone D and myrcene– tricyclene originated from South or Southwest Finland. Morphologically, the 20 tansy chemotypes based on the groups formed from complete linkage cluster analysis, were compared. The group containing the highest concentration of camphor chemotypes had the tallest shoots. The groups consisting from chemotypes containing davadone-D or artemisia ketone, which originated from Southwest Finland, produced the highest number of flower heads, had the tallest corymb, and were last to flower. Also, the group consisting from chemotypes with a high concentration of camphor and originated from South Finland started to flower late. The correlation between the genetic distance matrices based on RAPD patterns
*Corresponding author. Present address: Plant Production Research, Agricultural Research Center of Finland, FIN-31600 Jokioinen, Finland. Fax: +358-3-4188-2437.
E-mail address:[email protected] (M. Keskitalo).
reported previously (Keskitalo et al., 1998. Theor. Appl. Genet. 96, 1141–1150.) and the chemical distance matrices of the present study of the same tansy genotypes was highly significant (0.41,P50:0001). # 2001 Elsevier Science Ltd. All rights reserved.
Keywords:Asteraceae; Davadone-D; Essential oils; Intraspecific variation; Monoterpenes; Multivariate analysis; Petroleum-ether extraction;Tanacetum vulgareL.
1. Introduction
Tansy is an aromatic perennial plant, widely spread in the northern hemisphere (Heywood, 1976; Hussey, 1974; Hulte´n, 1968). Occasionally, it has been cultured in gardens (Mitich, 1992; Heywood, 1976) and used in salads, omeletts, cakes and spices (Grieve, 1984; Hussey, 1974). As a herbal remedy, tansy has traditionally been used in balsams, cosmetics, dyes, insecticides, medicines, and preservatives (Grieve, 1984; Hussey, 1974; Millspaugh, 1974) and as anthelmintic, for migrane, neuralgia, rheumatism and loss of appetite (Blumenthal, 1998), though the effectiveness for these uses have not been documented. Recent studies have also shown that the essential oil or extract of tansy exhibits anti-inflammatory (Brown et al., 1997; Mordujovich-Buschiazzo et al., 1996), antibactericidal (Neszme´lyi et al., 1992; He´thelyi et al., 1991; Holopainen and Kauppinen, 1989; Stefanovic et al., 1988), antifungicidal, (Neszme´lyi et al., 1992; He´thelyi et al., 1991), and a repellent against insects (Hough-Golstein and Hahn, 1992; Nottingham et al., 1991; Suomi et al., 1986; Panasiuk, 1984; Schearer, 1984). The activity against microbes and insects was dependent on the chemical composition of the essential oil (He´thelyi et al., 1991; Holopainen and Kauppinen, 1989; Panasiuk, 1984; Schearer, 1984). Components in the essential oil are also of potential interest as aroma chemicals in perfumery (Lawrence, 1992).
The composition of tansy oils varies markedly and several chemotypes from different geographical origins have already been classified. Tansy from Tierra del Fuego, Argentina (Gallino, 1988), and from England and USA (Ekundayo, 1979) were found to contain mainly b-thujone. Tansy oil obtained from two Canadian-grown plants contained b-thujone (d-isothujone) (Collin et al., 1993; von Rudolf and Underhill, 1965),a-thujone (l-thujone) (von Rudolf and Underhill, 1965), camphor-1,8-cineole-borneol, chrysanthenone, or dihydrocarvone (Collin et al., 1993) as the major oil components. In contrast, chemotypes such as artemisia alcohol, camphenol, davadone, lyratol, lyratyl acetate and 4-thujen-2a-yl (He´thelyi et al., 1981; Te´te´nyi et al., 1975), and trans-chrysanthenyl acetate (Neszme´lyi et al., 1992) have been detected in tansy grown in Hungary. Naturally occurring tansy from northeastern Netherlands contained artemisia ketone, chrysanthenol/chrysanthenyl acetate, lyratol/lyratyl acetate and b-thujone (Hendriks et al., 1990). Chrysanthenyl acetate and camphor/b-thujone chemotypes were identified in tansy grown in Belgium (De Pooter et al., 1989). In contrast, only one chemotype, chrysanthenyl acetate, was detected from tansy grown in
Piedmont, Italy (Nano et al., 1979). Several authors have studied the intraspecific variation of essential oil in Finnish tansy. Sorsa et al. (1968) found that a-pinene, b-pinene, 1,8-cineole, g-terpinene, artemisia ketone, thujone, camphor, umbellulone and borneol were the most common chemotypes. They also observed that camphor was the main component in northern, and thujone in southern Finland (Sorsa et al., 1968). Later, isopinocamphone (Forse´n and von Schantz, 1971),trans-chrysanthenyl acetate (Forse´n, 1974), sabinene, bornyl acetate, and germacrene D (Holopainen, 1989) chemotypes were detected from tansy grown in Finland.
Tansy also contains other secondary products such as less volatile sesquiterpenes (Chandra et al., 1987b) and several nonvolatile compounds. The latter group consists monoterpene glucosides (Banthorpe and Mann, 1971); sesquiterpene lactones (Sanz and Marco, 1991; Ognyanov and Todorova, 1983) such as parthenolide (Hendriks and Bos, 1990); sesquiterpene diketone (Ognyanov et al., 1983); triterpenes (Wilkomirski and Kucharska, 1992); and flavonoids (Chandra et al., 1987a; Ognyanov and Todorova, 1983).
In addition to different tansy chemotypes of which the concentration of one compound exceeds 40% (as a definition for well-defined chemotypes; Holopainen, 1989), more than 100 volatile mono- and sesquiterpenes have been identified from tansy by the authors including those compounds reported in this paper. Many factors are known to influence oil composition, including source of extractable plant tissue (leaf/flower), ontogeny of plant at sampling (Hendriks et al., 1990; Holopainen, 1989), seasonal changes (Ne´meth et al., 1994; von Schantz et al., 1966; von Rudolf and Underhill, 1965), and even the extraction method (Collin et al., 1993; Holopainen, 1989) may cause variation in the composition of volatile oil.
We are interested in to enhance the insecticidal properties of tansy. Thus, we have developed micropropagation, regeneration and protoplast fusion methods (Keskitalo et al., 1999, 1995) for the improvement of secondary metabolism by genetic engineering. We also studied genetic variation of 20 tansy genotypes collected from different regions of Finland by RAPD-PCR. Genetically, the genotypes clustered into two major groups, which were further divided into smaller groups, and the clustering was correlated to the geographical origin of the genotypes. Morphologically, the two clusters differed in their flowering (Keskitalo et al., 1998). To ascertain whether there is a connection between the chemical and genetic differences of tansy, we extracted the volatile compounds from the same tansy genotypes and analyzed these compounds with GC–MS. The goals of this study were two fold: (1) to identify the volatile terpenes extracted from air-dried tansy flower heads; and (2) to compare the variation to the geographical origin of the genotype to link the genetic and morphological variation among the tansy genotypes. Assessment of the chemical variation of tansy would allow us to identify unique chemotypes with potential industrial value and bioassay the activity of the genotypes with specific terpene composition. With such information the most desirable chemotypes can be selected for genetic engineering studies.
2. Materials and methods
2.1. Plant materials
The geographical locations of 20 tansy genotypes (Tv12Tv20) collected and used
in this study are listed in Table 1. The genotypes were transplanted to an orchard of the Department of Plant Production, University of Helsinki (608100N), Finland in
1991 (Keskitalo et al., 1998). The vouchers have been deposited at Botanical Museum of University of Helsinki (H). In the orchard, the plants were grown 1.5 m apart in two rows without fertilization. Morphological observations such as height of the stem, number of nodes per stem, number of flower heads per stem, length of corymb and date at the beginning of flowering were carried out as described (Keskitalo et al., 1998). The flower heads were excised at the onset of flowering, air dried at 388C, and stored in room temperature in the dark until extracted.
2.2. Isolation of volatile compounds
Two grams of air-dried flower heads crushed in a mortar, were transferred to centrifuge tubes containing 5 ml methanol, shaken for 1 h, centrifuged and the supernatant was collected to another tube. The flower heads were extracted with methanol 3 times. About 1–2 ml of saturated NaCl and 7 ml petroleum-ether (36–60)
Table 1
Geographical origins (National Landsurvey Institute, 1996) of the 20 tansy genotypes
Tansy genotype Collector # Keskitalo 21440
Region Latitude (N) Longitude (E) Voucher at (H)
Tv1 Iisalmi 638330N 278120E H 1698296
Tv2 Eno 628470N 308080E H 1698297
Tv3 Koli 638060N 298480E H 1698298
Tv4 Kangasniemi 618590N 268390E H 1698299
Tv5 Helsinki 608100N 248570E H 1698300
Tv6 Halikko 608230N 238030E H 1698301
Tv7 Lohjanharju 608300N 248240E H 1698302
Tv8 Tampere 618290N 238480E H 1698303
Tv9 Sa¨a¨ksma¨ki 618110N 248030E H 1698304
Tv10 Porvoo 608230N 258400E H 1698305
Tv11 Sipoo 608220N 258150E H 1698306
Tv12 Vihti 608250N 248180E H 1698307
Tv13 Ha¨meenlinna 608590N 248270E H 1698308
Tv14 Lemu 608330N 218590E H 1698309
Tv15 Tervakoski 608480N 248380E H 1698356
Tv16 Lieto 608300N 22827’E H 1698355
Tv17 Hanko 598490N 238000E H 1698354
Tv18 Alavus 628350N 2383700E H 1698353
Tv19 Hailuoto 658000N 248430E H 1698352
Tv20 Ko¨ylio¨ 618060N 228200E H 1698351
were added to the methanol extraction (15 ml), after which the solution separated into two phases. The top layer containing petroleum-ether was pipetted to an Erlenmeyer flask, and dried with anhydrous sodium sulfate in +48C in the dark for 1–3 days. The addition of NaCl and petroleum-ether was done twice. The dried supernatant was filtered and evaporated to 1 ml and stored in 2 ml vials in +48C in the dark until the GC analysis. The extraction from the flower heads of each tansy genotype was carried out 4 times. The different extraction times are referred as A–D.
2.3. Chemical analysis of the volatile compounds
Prior to the gas-chromatography (GC) analysis, 1ml of an internal standard (tetradecane; stock 50 mg/1 ml petroleum-ether) was added to a volume of 100ml of the extractions of A–C. The fourth sample (D) was run without the internal standard. A sample of 1ml was injected into the GC-Varian (Walnut Creek, CA) 3700 gas chromatograph, fitted with a flame-ionization detector and Hewlett-Packard model 3396 series II integrator (Palo Alto, CA). A fused silica capillary column (30 m0.25 mm i.d., 0.20mm film thickness, SPB-5; Supelco, Bellefonte, PA) was used. Helium was used as the carrier gas, and oven temperature was held isothermal at 808C for 5 min and then programmed to increase 38C/min to 2408C. The injector and detector temperatures were maintained at 180 and 2508C, respectively. GC/MS analyses were conducted using a Finnigan (San Jose, CA) GC (model 9610) and mass spectrometer (model 4000) system equipped with a capillary column under the same GC conditions and interfused to a Data General Nova/4 data processing system. The MS conditions included ionization voltage, 70 eV; emission current, 40 mA; scan rate, 1 scan/s; mass range, 40–500 Da; and ion source temperature, 1608C. The volatile components of the three separate extractions (A–C) in each tansy line were identified by comparing their relative retention times, retention indices (RI) and mass spectra with those of authentic samples, and by matching the mass spectra of each compound with different MS compound libraries for best fit (Adams, 1995; Finnigan et al., 1978; Stenhagen et al., 1974) and to the Kovats’ indices of authentic samples (Adams, 1995). The retention indices of our samples were calculated using n-alkanes (C8–C18) as a
reference (Table 2).
2.4. Statistical analysis
The volatile components of the tansy flower heads were scored as being present (1) or absent (0) in an extracted oil as determined by GC–MS. The pair-wise similarity and distance matrices of volatile compounds of the 20 tansy genotypes were calculated using the SAS program (SAS Institute, 1984) modified by Levy et al. (1991), similarly as we reported using RAPD-data (Keskitalo et al., 1998). The similarity and distance matrixes were calculated according the formula of Nei and Li (1979): [Sxy¼2Nxy:ðNxþNyÞ]; whereNxy is the number of terpenes common for
the accessionsxandy; andNxandNy are the numbers of terpenes of the accessions
xandy, respectively. The dendrogram was synthetized by complete linkage cluster
Table 2
Volatile oil concentration (%) and the mean of retention indices (RI) of volatile compounds extracted from air-dried flower heads of 20 tansy genotypes and identified by GC-MS. The results are a mean of three different extractions (A–C) carried out from each tansy genotype. The groups of tansy genotypes are formed according to the complete linkage cluster analysis based on the presence or absence of the identified compounda
Groups of tansy genotypes (Tv)
Group 1 Group 2 Group 3 Group 4 Group 5 Group 6
Compound RI Tv7 Tv8 Tv10 Tv20 Tv1 Tv3 Tv4 Tv18 Tv2 Tv5 Tv6 Tv12 Tv9 Tv11 Tv13 Tv15 Tv17 Tv19 Tv14 Tv16
1 Tricyclene 912 * 26.84 16.96 0.83 2.15 0.16 0.16 0.05 0.27 0.04 0.11 0.09 0.33 0.10 0.25 0.36 0.23 2.a-Thujene 932 0.04 0.36 0.54 0.34 0.17 0.39 0.28 0.56 0.30 0.73 0.53 0.45 0.07 3.a-Pinene 940 0.46 0.08 0.32 5.72 0.70 0.67 0.78 2.27 2.90 2.07 4.44 0.61 4.69 0.57 5.78 2.71 3.87 4. Camphene 957 2.44 0.64 2.42 6.03 7.27 0.61 5.15 7.76 8.66 12.60 4.57 11.62 5.06 2.24 7.83 0.60 0.42 5. Sabinene 980 0.10 0.04 2.14 1.62 0.03 0.28 0.22 2.95 0.85 0.03 0.29 0.40 1.61 0.72 0.64 1.12
6. Artemiseole 982 0.06
7.b-Pinene 986 0.08 0.85 0.05 2.64 0.61 0.15 2.01 1.03 1.46 2.57 0.60 2.91 0.40 1.27 0.80 1.46 0.04 8. Myrcene 1002 39.39 20.92 1.13 3.93
9. Yomogi alcohol 1002 2.74 1.69
10.r-Cymene 1029 1.43 1.22 2.04 0.04 1.61 2.91 3.42 1.77 0.22 1.32 1.96 1.02 4.00 1.22 2.05 1.39 1.71 1.57 0.68 0.43
11. Limonene 1033 0.75 1.95 0.06 1.36 0.16 0.34 2.53 0.38 1.20 0.81 0.37
12. 1,8-Cineole 1038 5.14 1.56 6.61 4.71 3.21 11.43 0.15 0.41 2.60 47.12 8.21 9.92 4.79 6.62 20.23 7.55 13.73 4.68 1.59 1.80 13. Artemisia ketone 1063 67.32 0.54 55.08 81.36 14.g-Terpinene 1063 0.31 1.39 0.54 1.24 1.18 0.41 0.31 0.78 0.21 0.76 0.45 0.70 0.40 0.86 0.64
15.cisSabinene hydrate
1079 0.02 0.75 0.34 0.55 0.19 0.08
16. Terpinolene 1097 1.56 0.88 1.53 0.27 0.78 0.82 0.65 0.24 0.18 0.14 0.69 0.58 0.18 0.19 1.24 0.39 0.67 0.28 17. Artemisia alcohol 1098 0.06 0.13 4.27 0.12 9.31 3.78 18.cisThujone 1116 0.86 0.07
19.transSabinene hydrate
1119 0.14 0.07
20. Chrysanthenone 1121 0.12 0.93 0.09 0.83 0.03
21.transThujone 1129 0.43 0.07 0.91 81.87 2.12 0.25 0.06 0.04 0.04 0.07
22. Camphor 1154 1.40 7.15 1.35 0.06 3.70 19.43 69.37 72.16 70.71 29.43 50.07 73.02 50.78 68.79 48.15 65.23 18.54 35.71 5.98 0.42 23. Chrysanthenol 1164 0.69 3.07 2.00 1.12
24. Pinocarvone 1168 4.38 1.06
25. Umbellulone 1170 12.18 0.15 4.94 2.65 2.76 1.01 0.56 0.12 0.65 0.03 3.95 0.39 3.73 2.46 1.53
26. Terpinen-4-ol 1175 0.66 0.99
27. Artemisyl acetate 1176 4.35 15.26 2.45 1.73 3.45 3.01 5.81 0.11 1.39 0.60 4.17 5.01 1.58
28. Borneol 1177 0.81 3.87 5.08 0.34
29.cisPinocamphone 1185 0.04 0.03 0.04 0.03 19.10
30.a-Terpineol 1199 0.30 1.16 1.11 0.07 0.16 4.27 0.65 1.01 0.62 0.33 0.48
31. Myrtenol 1204 10.69
Groups of tansy genotypes (Tv)
Group 1 Group 2 Group 3 Group 4 Group 5 Group 6
Compound RI Tv7 Tv8 Tv10 Tv20 Tv1 Tv3 Tv4 Tv18 Tv2 Tv5 Tv6 Tv12 Tv9 Tv11 Tv13 Tv15 Tv17 Tv19 Tv14 Tv16
35. Carvone 1250 0.97 0.65 0.81 0.90 0.81 36.cisChrysanthenyl
acetate
1271 0.89 4.39 11.97 24.88 0.05 1.42
37. Unknown, MW 154
1286 0.07
38. Bornyl acetate 1291 0.67 0.54 0.43 5.16 0.81 0.11 11.55 0.04 17.80 1.16 0.31 1.35 5.72 16.72 0.74 39. Isobornyl acetate 1298 0.06 12.04 7.48
40. Carvacrol 1304 0.06 2.67
41. Thymol 1306 0.58 1.37 0.86
42.d-Elemene 1336 0.65 0.66
43.b-Cubebene 1387 0.07 0.26 0.40
44.b-Elemene 1390 2.25 0.15 0.19 0.07 0.13 0.58 0.52 0.14 0.07 0.11 0.86 0.03 1.38 0.09 0.28 45. (E) Caryophyllene 1426 0.90 1.36 2.05 0.77 1.43 0.96 0.23 1.27 0.04 1.32 0.53 0.88 0.43 0.43 0.59 1.31 1.32 0.86 46. Germacrene D 1485 1.58 2.54 0.75 1.56 0.66 2.18 2.66 1.14 0.82 0.79 5.08 1.08 0.03 0.99 1.47 0.44 1.12 0.12 47. Bicycloelemene 1501 1.32 0.06 0.60 0.05
48.g-Cadinene 1525 2.64 1.87 0.86 2.92 0.39 1.13 1.03 49. Artedouglasia
oxide C
1553 1.18
50. Nerolidol 1565 2.22 0.37 0.04 3.11 1.27 1.74 0.26
51. Germacrene
1594 0.55 0.17 0.19 0.13 0.41 0.25 0.16
54. Davadone D 1597 65.51 0.55 1.81 55. Unknown,
MW 222
1623 0.39 0.13 0.33 0.15 0.09 0.23
Percentage of total peak area identified
85.54 95.85 95.61 98.05 98.70 98.55 97.76 98.89 97.00 98.64 98.91 99.37 99.29 98.68 99.33 99.11 97.79 96.84 97.94 97.39
Total amount (mg) of terpenes isolated from 2.0 g flower heads
0.93 1.61 1.11 5.46 2.84 2.36 3.29 2.29 1.03 2.80 1.50 2.73 1.94 3.45 5.07 5.65 2.22 4.35 4.76 3.20
aBlank areas = not detected
analysis, which illustrates the chemical distance as the normalized maximum distance between clusters. The normalized maximum distance is a normalized value (distance betweenxandy/mean distance of the population) of the average genetic distance.
Morphological data of tansy genotypes (Keskitalo et al., 1998) were arranged according to the grouping of complete linkage cluster analysis and the differences between the groups were calculated with SAS (SAS Institute, 1984) CONTRAST program. To compare the distance matrices of RAPD-PCR study of our previous study (Keskitalo et al., 1998) and the distance matrices of this present study (Table 3), Pearson correlation coefficient was calculated with SAS.
3. Results
3.1. Volatile compounds identified with GC–MS
A total of 55 aromatic volatile compounds were detected from the petroleum-ether extraction of dried flower heads of tansy. The volatile compounds covered on average 97.46% (S.D. 2.69) of the total peak area recovered from GC. The con-centration of 47 of the 55 compounds varied highly significantly between the tansy genotypes (P50:0001) and only the genotypic variation in artemiseole, trans
sabi-nene hydrate, nerol, nerolidol, spathulenol, caryophyllene oxide and two unidenti-fied compound was not significant. On average, the total amount of volatile com-pounds extracted from 2.0 g of tansy flower heads was 2.90 mg (S.D. 1.63) (Table 2). Fifteen of the 20 genotypes had main component (usually camphor, artemisia ketone,
transthujone, or davadone D), which consisted at least 40% of the total peak area, and five genotypes contained at least two terpenes as the main components.
3.2. Complete linkage cluster analysis
The smallest chemical distance observed was 0.116 between the chemotypesTv5 and Tv 6 and the largest distance between Tv 7 and Tv 19 (Table 3). The mean chemical distance was 0.374 among the entire tansy population. Complete linkage cluster analysis separated the population to six groups. The largest and smallest normalized maximum distance was 1.94 (0.726/0.374=1.94) and 0.31 (0.116/ 0.371=0.31) between group 1 and the cluster of the other groups, and betweenTv
5 andTv 6, respectively (Fig. 1).
3.3. Volatile compounds related to the geographical origin, RAPD-PCR pattern and morphology
Six of the seven chemotypes, which did not contain camphor as the main component, originated from Southern Finland (Tv7, 8, 10, 14, 16, 20), and only the chemotype containing thujone (Tv1) was from Central Finland (Fig. 2). Eight of the 13 chemotypes with camphor concentration exceeding 18.5% originated from Central Finland (Tv2, 3, 4, 9, 13, 15, 18, 19) and five from Southern Finland (Tv5, 6,
11, 12, 17). Artemisia ketone (Tv 14, 16, 20) was found only from genotypes originating from Southwestern Finland. Also, geographically the six groups formed from complete linkage cluster analysis could be separated from each other. Only the groups containing a high concentration of davadone D, artemisia ketone or myrcene-tricyclene (Tv7, 8, 10, 20 andTv 14, 16) were found from South-Finland. The groups containing a high concentration of camphor were found from South and Central Finland, but the number of chemotypes containing a high concentration of camphor was higher in Central Finland (Fig. 2).
In a previous paper, we reported the genetic distance matrices based on RAPD-PCR patterns (Keskitalo et al., 1998). The groupings of tansy population according to our previous genetic study (Keskitalo et al., 1998) and this study on tansy volatile compounds are illustrated in Fig. 2. Because the distance matrices of these both studies were calculated by the same method, we calculated the correlation between matrices of genetic data and chemical data. The Pearson correlation coefficient, 0.407 (P50:0001), revealed a 16.565% analogy between genetic and chemical
differences based on absence or presence of compound in the 20 tansy genotypes studied.
The tansy genotypes were arranged according to the six groups formed from complete linkage cluster analysis and the variation of morphology was compared between the groups using the SAS CONTRAST procedure. The group containing the highest percentage of camphor and 1,8-cineole (Tv 9, 11, 13, 15) had the tallest shoots (108.2 cm) and differed significantly from the others (P50:0001) whereas the Fig. 1. The clustering based on the absence or presence of 55 volatile compounds identified by GC–MS of the 20 tansy genotypes used in this study. The chemical distance between genotypes were calculated as described by Nei and Li (1979) using the complete linkage cluster analysis. The bar shows the normalized maximum distance between clusters.
Table 3
Chemical distance of the 20 tansy genotypes based on the absence or presence of volatile compound using the method of Nei and Li (1979)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Tv1 0.000
Tv2 0.351 0.000
Tv3 0.156 0.368 0.000
Tv4 0.217 0.384 0.234 0.000
Tv5 0.200 0.211 0.217 0.234 0.000
Tv6 0.190 0.200 0.163 0.227 0.116 0.000
Tv7 0.514 0.600 0.579 0.487 0.632 0.600 0.000
Tv8 0.348 0.487 0.362 0.417 0.362 0.318 0.590 0.000
Tv9 0.292 0.366 0.265 0.200 0.184 0.217 0.512 0.360 0.000
Tv10 0.347 0.476 0.360 0.333 0.400 0.319 0.524 0.255 0.283 0.000
Tv11 0.333 0.415 0.388 0.200 0.265 0.304 0.561 0.480 0.269 0.358 0.000
Tv 12 0.282 0.250 0.300 0.317 0.150 0.135 0.688 0.415 0.265 0.409 0.256 0.000
Tv13 0.261 0.333 0.319 0.333 0.149 0.182 0.641 0.375 0.200 0.373 0.280 0.220 0.000
Tv14 0.463 0.412 0.524 0.535 0.429 0.385 0.588 0.488 0.422 0.478 0.511 0.444 0.349 0.000
Tv15 0.265 0.333 0.280 0.255 0.200 0.234 0.619 0.451 0.245 0.407 0.245 0.318 0.216 0.391 0.000
Tv16 0.422 0.421 0.435 0.362 0.304 0.349 0.526 0.489 0.224 0.480 0.388 0.400 0.319 0.286 0.320 0.000
Tv17 0.362 0.500 0.375 0.388 0.375 0.378 0.600 0.388 0.373 0.385 0.412 0.476 0.224 0.409 0.308 0.500 0.000
Tv18 0.256 0.278 0.273 0.156 0.182 0.171 0.556 0.378 0.277 0.375 0.277 0.263 0.289 0.450 0.208 0.318 0.391 0.000
Tv19 0.364 0.405 0.333 0.348 0.333 0.238 0.730 0.435 0.417 0.469 0.417 0.385 0.348 0.512 0.265 0.511 0.362 0.256 0.000
Tv20 0.526 0.484 0.538 0.600 0.538 0.444 0.613 0.400 0.571 0.442 0.619 0.636 0.550 0.543 0.488 0.538 0.561 0.514 0.474 0.000
M.
Keskitalo
et
al.
/
Biochemical
Systematics
and
Ecology
29
(2001)
267–285
group containing mixed chemotypes (Tv 17, 19) had the shortest shoots (72.7 cm). The group having the shortest shoots (Tv17, 19) and tallests shoots (Tv9, 11, 13, 15) had the same number of nodes (20.9) in the stems, the number which was the highest and differed significantly (P50:0001) from the number of nodes in the other groups.
The number of flower heads in the groups, which contained davadone D/myrcene– tricyclene/artemisia ketone (Tv7, 8, 10, 20; 50.6), artemisia ketone (Tv14, 16; 56.5) or the highest concentration of camphor (Tv9, 11, 13, 15: 50.9) was the highest, and the results differed significantly (P50:005) from the other groups. The mixed group containing pinocamphone-camphor-1,8-cineole-bornyl acetate (Tv 17, 19) had the lowest number of flower heads (20.9) per stem and differed significantly (P50:001)
from the other groups. Flowering was latest in the groups containing the lowest concentration camphor (Tv 7, 8, 10, 20 and Tv 14, 16) or a high concentration of camphor (Tv 2, 5, 6, 12), and the date at the beginning of the flowering differed significantly (P50:05) from the other three groups (Table 4).
Tansy chemotypes could be joined to two clusters based on camphor concentra-tion. The cluster consisting of chemotypes which camphor concentration was between 0.1 and 7.2% (Tv 1, 7, 8, 10, 14, 16, 20) had less nodes per stem (average 16.6; P50:001), more flower heads (average 50.4; P50:001) and a taller corymb
(average 13.9 cm; P50:01), than the other cluster consisting of chemotypes which
camphor concentration was more than 18.5% (Tv 2, 3, 4, 5, 6, 9, 11, 12, 13, 15, 17, 18, 19). The number of nodes, number of flower heads and the height of the corymb of the latter cluster were on average 18.8, 41.2 and 11.1 cm, respectively (data not shown).
4. Discussion
In total, 55 volatile compounds were detected from air-dried tansy flower heads of which 53 were identified. The most frequently found compounds exceeding 10% at least in one chemotype were tricyclene, camphene, myrcene, 1,8-cineole, artemisia ketone, trans thujone, camphor, umbellulone, artemisyl acetate, pinocamphone, myrtenol, chrysanthenyl acetate, bornyl acetate, and davadone D, which have already reported in tansy (Collin et al., 1993; Neszme´lyi et al., 1992; Hendriks et al., 1990; De Pooter et al., 1989; Gallino, 1988; Holopainen, 1989; He´thelyi et al., 1981; Ekundayo, 1979; Nano et al., 1979; Te´te´nyi et al., 1975; Forse´n, 1975, 1974; Forse´n and von Schantz, 1971; Sorsa et al., 1968; von Rudolf and Underhill, 1965), or from related species such as artemisyl acetate fromArtemisiaspp. (Worku and Rubiolo, 1996; Epstein and Gaudioso, 1984). An artemisia ketone isomer observed previously in tansy oil (Hendriks et al., 1990) shows a retention index linear with our observation of artemisyl acetate. Most minor compounds reported here have been reported in tansy, but seven minor compounds are reported here for the first time. All seven have been detected from related species in Asteraceae family. These include artemiseole from Artemisia arbuscula (Epstein and Gaudioso, 1984), isobornyl acetate from Artemisia vulgaris (Hwang et al., 1985), artedouglasia oxide from Artemisia laciniata (Weyerstahl et al., 1997), nerolidol from Tanacetum cinerariifolium, Artemisia lacinata,andTanacetum polycephalum(Saggar et al., 1997;
Weyerstahl et al., 1997; Rustaiyan et al., 1990, respectively), germacrene alcohol fromEchinacea purpurea(Bauer et al., 1988), spathulenol fromAchillea millefolium
and Achilla laciniata (Afsharypuor et al., 1996a; Weyerstahl et al., 1997), and caryophyllene oxide from Tanacetum annuum, Achillea wilhelmsii and Achillea laciniata(Barrero et al., 1992; Afsharypuor et al., 1996b; Weyerstahl et al., 1997, respectively). Interestingly, davadone D which has previously been found only in
Fig. 2. (a) The Geographical origins of the 20 tansy genotypes used in this study. The groups 1 and 2 are defined on the basis of a complete linkage cluster analysis of RAPD-PCR data (Keskitalo et al., 1998). (b) The groups of 1–6 are defined on the basis of a complete linkage cluster analysis of the presence or absence of volatile compounds analyzed in this study.
Hungarian-grown (Ne´meth et al., 1994; He´thelyi et al., 1991, 1981), was detected here for the first time in tansy grown in Finland.
Among the 20 tansy genotypes, 15 had a major compound, the concentration of which exceeding 40%, while the oil of five genotypes was composed of several minor compounds. Based upon the definition of ‘chemotypes’ by previous authors (Hendriks et al., 1990; Holopainen, 1989), 15 tansy genotypes are ‘well-defined’ chemotypes, and the remaining five are ‘mixed’ chemotypes. Holopainen (1989) observed that about 20% of tansy chemotypes resulting from their crossing experi-ments were also mixed chemotypes. In our study, camphor was the most common
Fig. 2 (continued.)
Table 4
Morphological attributes of tansy genotypes clustered into six groups according to the complete linkage cluster analysis based on the presence or absence of the identified compound
Main compound(s) of
2. Camphor; 4,18; 69.4–72.2; 41.2 34.8 93.7 9.2 16.9 1.1 40.0 15.8 11.3 4.3 22nd July 2.9 Thujone; 1; 81.9;
Chrysanthenyl acetate/camphor/1,8-cineole; 3; 24.9/19.4/11.4;
3. Camphor; 2, 6,12; 50.1–73.0; 55.8 20.4 84.4 18.5 17.3 3.2 40.2 8.6 10.0 1.8 30th July 4.2 1,8-Cineole/camphor; 5; 47.1/29.4;
4. Camphor; 9,11,15; 50.8–68.9; 58.2 10.3 108.2 12.0 20.9 2.8 50.9 5.7 12.5 2.6 24th July 4.4 Camphor/1,8-cineole; 13; 48.2/20.2;
5. Pinocamphone/camphor/1,8-cineole; 17; 19.1/18.5/13.7; 27.1 12.1 72.7 17.3 20.9 8.2 24.0 9.0 8.7 5.3 26th July 4.9 Camphor/bornyl acetate; 19; 35.7/16.7;
6. Artemisia ketone; 14,16; 55.1–81.4; 3.2 3.9 89.6 17.2 15.0 0.6 56.5 33.0 18.8 8.4 30th July 0.5
aMain compound(s) consisting at least 10% of the total peak area of each chemotype.
bTansy chemotype(s) consisting the compound.
cConcentration of camphor in the group.
chemotype, and this is in agreement with others (Hendriks et al., 1990; Holopainen, 1989; Ekundayo, 1979; Sorsa et al., 1968). In mixed camphor chemotypes, camp-hene, 1,8-cineole, pinocampcamp-hene, chrysanthenyl acetate, bornyl acetate, and iso-bornyl acetate were the most frequently found associated compounds (10–30%). We also identified three well-defined artemisia ketone chemotypes. Such chemotypes have been observed previously by Forse´n and von Schantz (1971) and Sorsa et al. (1968) from Finnish tansy, though are common from the Netherlands and Hungary (Hendriks et al., 1990; Te´te´nyi et al., 1975). A davadone D chemotype, which is reported now for the first time from tansy grown in Finland, was detected only from a Hungarian-grown tansy (Ne´meth et al., 1994; He´thelyi et al., 1991, 1981). In contrast to the previous authors (Holopainen, 1989; Forse´n, 1975; Sorsa et al., 1968) who studied essential oil of Finnish tansy, thujone was not among the most common monoterpenes in our study. Thujone was found only in small concentra-tions except for one chemotype where it was the main component. Our observation was more in accordance with Hendriks et al. (1990), who did not detect high concentration of thujone together with camphor. The mixed chemotypes containing tricyclene (16–27%) and myrcene (21–39%) have not been reported previously from tansy in Finland, although Sorsa et al. (1968) found chemotypes containing eithera -pinene or tricyclene (4–14%) accompanied with a low concentration (0.1–0.3%) of myrcene.
Geographically, most of the chemotypes containing a significant concentration of camphor originated from Central Finland, whereas chemotypes containing camphor as a minor compound originated from South or Southwest Finland. This observation is in agreement with Sorsa et al. (1968) who found that camphor was more frequently observed in tansy grown in Northern Finland compared to southern grown tansy, where thujone was more frequent. Artemisia ketone and davadone D were detected only from tansy originating from Southwest Finland. Similar geographical variation in terpene composition and chemotypes of tansy have been observed within and between other countries (Neszme´lyi et al., 1992; Hendriks et al., 1990; De Pooter et al., 1989; Gallino, 1988; He´thelyi et al., 1981 Ekundayo, 1979; Nano et al., 1979; Te´te´nyi et al., 1975; von Rudolf and Underhill, 1965).
In Finland, the genetic variation between the genotypes originating from different geographical regions may be the result of naturalization through inhabitation and agriculture, with the adaptation of tansy to the local climate (Keskitalo et al., 1998). Tansy has been observed to be one of the most common seed species in the ballast soil area in Reposaari (Jutila, 1996), which has been an important harbor in Southwest Finland. Many plant species have spread to Finland by seed embedded in the ballast soils used in ships (Jutila, 1996). Interestingly, tansy originating from Southwest Finland contained only artemisia ketone or davadone D as the main com-ponents, common compounds observed in tansy from the Netherlands (Hendriks et al., 1990) and Hungary (He´thelyi et al., 1981). Correlation between the genetic distance matrices of our previous study (Keskitalo et al., 1998) and the chemical distance matrices of the present study was 0.407 showing some analogy between the variation of the two matrices. Unfortunately, only in a few cases has terpene
variation been studied simultaneously with genetic variation. In agreement with our study,Juniperusspecies have been successfully differentiated from each other using volatile terpenoids analysis and RAPD patterns (Adams et al., 1993).
The group containing camphor and 1,8-cineole, had the tallest shoots while the mixed chemotypes were the shortest. Ne´meth et al. (1994) observed that chemo-types containing a thujen-acetate had the tallest shoots whereas plants with 1,8-cineole were the shortest. We also observed that the group containing davadone D were among the groups having a highest number of flower heads per shoots, which is in agreement with the observation of Ne´meth et al. (1994). We observed, that groups formed from pure chemotypes produced the highest number of flower heads whereas groups formed from mixed chemotypes had a low number of flower heads. Overall, in our study, the chemotypes containing a high concentration of camphor had less flower heads and initiated flowering earlier than chemotypes low in camphor. This is in agreement with observations by Ne´meth et al. (1994).
The underlying causes of the observed chemical variation in tansy is an intriguing question, and the answer still remains elusive. Genetic variation may be due to the different geographical origins of tansy (Keskitalo et al., 1998), which eventually led to differences in the genetic control of essential oil accumulation (Holopainen, 1989; Lokki et al., 1973). A wide variation in essential oil composition presumably has ecological advantages in protecting plants against different pests (Hough-Golstein and Hahn, 1992; Neszme´lyi et al., 1992; He´thelyi et al., 1991; Nottingham et al., 1991). It is also likely, that part of the terpene expression in tansy is linked to specific environmental or climatic conditions (Sorsa et al., 1968), and to a lesser extent may be an indicator of other characters such as morphology (Ne´meth et al., 1994).
The dependency between the geographical origin, genetic, chemical and morphological variation of tansy show that different factors need to be recognized when the biodiversity of herbaceous species is to be examined. The analogy between the relative chemical and genetic differences among the 20 tansy chemo- and genotypes, respectively, suggest that different terpene compositions resulting from the differential activation of specific enzymes may be related to the variation in RAPDs patterns. The association between the main chemical components and morphology should be considered when selecting the parental chemo- and phenotypes from tansy populations for future work. The use of morphological traits as indicators of selected chemotypes would be most useful in breeding and biochemical studies. Since the bioactivity of the essential oil of tansy depends on the composition of terpenes (He´thelyi et al., 1991; Holopainen and Kauppinen, 1989; Panasiuk, 1984; Schearer, 1984), the chemotype with the most effective oil composition should be selected. According to previous studies of bioactivity of tansy oil, artemisia ketone (He´thelyi et al., 1981), camphor (Holopainen and Kauppinen, 1989; Schearer, 1984), chrysanthenyl acetate (Neszme´lyi et al., 1992), 1,8-cineole (Schearer, 1984), davadone (He´thelyi et al., 1981), and thujone (Holopainen and Kauppinen, 1989) chemotypes are among the most interesting ones for further studies..
Acknowledgements
The authors thank Professor M. Levy, Department of Biological Sciences, Purdue University for providing the SAS-supported complete linkage cluster program. Financial support from the Academy of Finland (grant 7798), Finnish Association of Academic Agronomists (Agronomiliitto), and the Rotary Foundation of Southwest Finland (district 1410) is gratefully acknowledged.
References
Adams, R., 1995. Identification of Essential Oil Components by Gas Chromatography/Mass Spectro-scopy. Allured Publishing Corporation, IL USA, 469 p.
Adams, R.P., Demeke, T., Abulfatih, H.A., 1993. RAPD DNA fingerprints and terpenoids: clues to past migrations ofJuniperusin Arabia and East Africa. Theor. Appl. Genet. 87, 22–26.
Afsharypuor, S., Asgary, S., Lockwood, G.B., 1996a. Volatile constituents ofAchillea millefoliumL. ssp millefoilumfrom Iran. Flavour Fragrance J. 11, 265–267.
Afsharypuor, S., Asgary, S., Lockwood, G.B., 1996b. Constituents of the essential oil of Achillea wilhelmsiifrom Iran. Planta Med 62, 77–78.
Banthorpe, D.V., Mann, J., 1971. Monoterpene glucosides from petals of Tanacetum vulgare. Phytochemistry 11, 2589–2591.
Barrero, A.F., Sa´nchez, J.F., Altarejos, J., Zafra, M.J., 1992. Homoditerpenes from the essential oil of Tanacetum annuum. Phytochemistry 31, 1727–1730.
Bauer, R., Remiger, P., Wray, V., Wagner, H., 1988. A germacrene alcohol from fresh aerial parts of Echinacea purpurea. Planta Med. 54, 478–479.
Blumenthal, M. (Translator), 1998. The Complete German Comission E Monographs: Therapeutic Guide to Herbal Medicines. Tansy Flower and Herb. Unapproved Herbs. American Botanical Council, Austin, TX, Integrative Medicine Communications, Boston, MA, pp. 379–380.
Brown, A.M.G., Edwards, C.M., Davey, M.R., Power, J.B., Lowe, K.C., 1997. Effects of extracts of Tanacetum species on human polymorphonuclear leucocyte activity in vitro. Phytothe. Res. 11, 479–484.
Chandra, A., Misra, L.N., Thakur, R.S., 1987a. Germacranolides and an alkyl glucoside fromTanacetum vulgare. Phytochemistry 26, 1463–1465.
Chandra, A., Misra, L.N., Thakur, R.S., 1987b. Tanavulgarol, an oxygenated sesquiterpene with an uncommon skeleton fromTanacetum vulgare. Phytochemistry 26, 3077–3078.
Collin, G.J., Deslauriers, H., Pageau, N., Gagnon, M., 1993. Essential oil of tansy (Tanacetum vulgareL.) of Canadian origin. J. Essent. Oil Res. 5, 629–638.
Ekundayo, O., 1979. Essential oils II. Terpene composition of the leaf oil ofTanacetum vulfgareL. Z. Pflanzenphysiol. Bd. 92, 215–220.
Epstein, W.W., Gaudioso, L.A., 1984. Volatile oil constituents of sagebrush. Phytochemistry 23, 2257–2262.
Finnigan, Heller, Milne, 1978. 4000 GC/MS; NBS Library 31, 331 entries
Forse´n, K., 1974. Begleitstoffe in verschiedenen Chemotypen vonChrysanthemum vulgareReine trans-Chrysanthenylacetate-type. Farm. Aikak 83, 9–17.
Forse´n, K., 1975. U¨ber die Infraspezifische chemische variation beiChrysanthemum vulgare. Ann. Acad. Sci. Fenn. Ser. A IV 207, 1–54.
Forse´n, K., von Schantz, M., 1971. Neue Hauptbestandteile im a¨therischen o¨l des Rainfarns in Finnland. Arch. Pharm. 304, 944–952.
Gallino, M., 1988. Essential oil fromTanacetum vulgaregrowing spontaneously in ‘Tierra del Fuego’ (Argentina). Planta Med. 54, 182.
Grieve, M., 1984. Tansy. In: Leyel, C.F. (Ed.), A Modern Herbal. Penguin Books Ltd, Middlesex, Great Britain, pp. 789–790.
Hendriks, H., Bos, R., 1990. The presence of parthenolide in Dutch tansy. Planta Med. 56, 550. Hendriks, H., van der Elst, D.J.D., van Putten, F.M.S., Bos, R., 1990. The essential oil of Dutch Tansy
(Tanacetum vulgareL.). J. Essent. Oil Res. 2, 155–162.
Heywood, V.H., 1976.TanacetumL. In: Tutin, T.G., Heywood, V.H., Burges, N.A., Moore, D.M., Valentine, D.H., Walters, S.M., Webb, D.A. (Eds.), Flora Europaea, Plantaginaceae to Compositae (and Rubiaceae), Vol. 4. Cambridge University Press, Cambridge, Great Britain, pp. 169–171. He´thelyi, E´., Te´te´nyi, P., Danos, B., Koczka, I., 1991. Phytochemical and antimicrobial studies on the
essential oils of theTanacetum vulgareclones by gas chromatography/mass spectrometry. Herba Hung 30, 82–90.
He´thelyi, E´., Te´te´nyi, P., Kettens-van den Bosch, J.J., Salemink, C.A., Heerma, W., Versluis, C., Kloosterman, J., Sipma, G., 1981. Essential oils of fiveTanacetum vulgaregenotypes. Phytochemistry 20, 1847–1850.
Holopainen, M. (1989). A study on the essential oil of tansy (Tanacetum vulgare L.). Ph.D. Thesis, Department of Pharmacy, University of Helsinki, Finland.
Holopainen, M., Kauppinen, V., 1989. Antimicrobial activity of different chemotypes of Finnish tansy. Planta Med 55, 103.
Hough-Golstein, J., Hahn, S.P., 1992. Antifeedant and oviposition deterrent activity of an aqueous extract ofTanacetum vulgareL. on two cabbage pest. Environ. Entomol. 21, 837–844.
Hulte´n, E., 1968. Flora of Alaska and Neighboring Territories. A Manual of the Vascular Plants. Stanford University Press, Stanford CA, 1008 p.
Hussey, J., 1974. Some useful plants of early New England. Econ. Bot. 28, 311–337.
Hwang, Y.S., Wu, K.H., Kumamoto, J., Axelrod, H., Mulla, M.S., 1985. Isolation and identification of mosquito repellents inArtemisia vulgaris. J. Chem. Ecol. 11, 1297–1306.
Jutila, H.M., 1996. Seed bank and emergent vascular flora of ballast areas in Reposaari, Finland. Ann. Bot. Fennici. 33, 165–182.
Keskitalo, M., Angers, P., Earle, E., Pehu, E., 1999. Chemical and genetical characterization of protoplast fusion-derived calli between tansy (Tanacetum vulgareL.) and pyrethrum (Tanacetum cinerarieafolium). Theor. Appl. Genet. 98, 1335–1343.
Keskitalo, M., Kanerva, T., Pehu, E., 1995. Development of in vitro procedures for regeneration of petiole and leaf explants and production of protoplast-derived callus inTanacetum vulgareL. (Tansy). Plant Cell Rep. 14, 261–266.
Keskitalo, M., Linde´n, A., Valkonen, J.P.T., 1998. Genetic and morphological diversity of Finnish tansy (Tanacetum vulgareL.). Theor. Appl. Genet. 96, 1141–1150.
Lawrence, B.M., 1992. Essential oils as source of natural aroma chemicals. Perfumer Flavorist 17, 15–28. Levy, M., Romao, J., Marcetti, M.A., Hamer, J.E., 1991. DNA Fingerprinting with a dispersed repeated
sequence resolves pathotype diversity in rice blast fungus. Plant Cell 3, 95–102.
Lokki, J., Sorsa, M., Forse´n, K., von Schantz, M., 1973. Genetics of monoterpenes inChrysanthemum vulgare. Hereditas 74, 225–232.
Millspaugh, C.F., 1974. American Medicinal Plants. An Illustrated and Descriptive Guide to Plants Indigenous to and Naturalized in the United States which are used in Medicine. Dover Publications, New York, 806p.
Mitich, L.W., 1992. Intriguing world of weeds}Tansy. Weed Technol. 6, 242–244.
Mordujovich-Buschiazzo, P., Balsa, E.M., Busciazzo, H.O., Mandrile, E., Rosella, M., Schinella, G., Fioravanti, D., 1996. Anti-inflammatory activity of Tanacetum vulgare. Fitoterapia LXVII, 319–322.
Nano, G.M., Bicchi, C., Frattini, C., Gallino, M., 1979. Wild Piedmontese Plants II. A rare chemotype of Tanacetum vulgareL., abundant in Piedmont (Italy). Planta Med. 35, 270–274.
National Landsurvey Institute, 1996. Geographic Data Center.
Ne´meth, E´.Z., He´thelyi, E´., Berna`th, J., 1994. Comparison studies onTanacetum vulgareL. chemotypes. J. Herb Spices Med. Plants 2, 85–92.
Neszme´lyi, A., Milne, G.W.A., He´thelyi, E´., 1992. Composition of the essential oil of clone 409 of Tanacetum vulgareand 2D NMR investigation oftrans-chrysanthenyl acetate. J. Essent. Oil Res. 4, 243–250.
Nei, M., Li, W.H., 1979. Mathematical model for study genetic variation in terms of restriction endonucleases. Proc. Natl. Acad. Sci. USA 74, 5267–5273.
Nottingham, S.F., Hardie, J., Dawson, G.W., Hick, A.J., Picett, J.A., Wadhams, L.J., Woodcock, C.M., 1991. Behavioral and electrophysiological responses of aphids to host and nonhost plant volatiles. J. Chem. Ecol. 17, 1231–1242.
Ognyanov, I., Todorova, M., 1983. Sesquiterpene lactones and flavonoids in flowers ofTanacetum vulgare. Planta Med. 48, 181–183.
Ognyanov, I., Todorova, M., Dimitrov, V., Ladd, J., Irngartinger, H., Kurda, E., Rodewald, H., 1983. Cis-Longipinane-2,7-dione, a sesquiterpene diketone in flowers ofTanacetum vulgare. Phytochemistry 22, 1775–1777.
Panasiuk, O., 1984. Response of Colorado potato beetles,Leptinotarsa decemlineata(Say), to volatile components of tansy,Tanacetum vulgare. J. Chem. Ecol. 10, 1325–1333.
De Pooter, H.L., Vermeesch, J., Schamp, N.M., 1989. The essential oils ofTanacetum vulgareL. and Tanacetum parthenium(L.) Schultz-Bip. J. Essent. Oil Res. 1, 9–13.
von Rudloff, E., Underhill, E.W., 1965. Gas-liquid chromatography of terpenes}XII. Seasonal variation
in the volatile oil fromTanacetum vulgareL. Phytochemistry 4, 11–17.
Rustaiyan, A., Zare, K., Habibi, Z., Hashemi, M., 1990. Germacranolides fromTanacetum polycephalum. Phytochemistry 29, 3022–3023.
Saggar, P., Wamicha, W.N., Chhabra, S.C., Ndalut, P., 1997. Isolation, identification and bioassay of repellent factors in the essential oil of pyrethrum for grain protection against Sitophilus zeamais (Molts.). Pyrethrum Post 19 (4), 126–130.
Sanz, J.F., Marco, J.A., 1991. NMR studies of tatridin A and some related sesquiterpene lactones from Tanacetum vulgare. J. Nat. Prod. 54, 591–596.
SAS Institute Inc., 1984. SAS/STAT Users Guide. Version 6, 4th Edition. 1686p.
von Schantz, M., Ja¨rvi, M., Kaartinen, R., 1966. Die Vera¨nderungen des A¨therischen O¨les wa¨hrend der Entwicklung der Blu¨tenko¨rbchen vonChrysanthemum vulgareL. Planta Med. 14, 421–435.
Schearer, W.R., 1984. Components of oil of tansy (Tanacetum vulgare) that repel Colorado potato beetles (Leptinotarsa decemlineara). J. Nat. Prod. 47, 964–969.
Sorsa, M., von Schantz, M., Lokki, J., Forse´n, K. (1968). Variability of essential oil components in Chrysanthemum vulgareL. in Finland. Ann. Acad. Sci. Fenn. Ser. A VI, 135, 1-13.
Suomi, D., Brown, J.J., Akre, R.D., 1986. Responses to plant extracts of Neonatal codling moth larvae, Cydia pomonella(L.), (Lepidoptera: Tortricidae: Olethreutinae). J. Entomol. Soc. B. C. 83, 12–18. Stefanovic, M., Ristic, N., Vukmirovic, M., 1988. Biological activities of sesquiterpene lactones.
Investigations of microbial activities of lactones isolated from the Yugoslav plant species of the genus TanacetumL. [Chrysanthemum (fam. Compositae)]. Bull. T. XCV Acad. Serbe Sci. Arts, Classe Sci. Naturelles Math. Sci. Naturelles 28, 23–43.
Stenhagen, E., Abrahamssons, S., McLafferty, F., 1974. Registry of Mass Spectral Data. Wiley, New York.
Te´te´nyi, P., Kaposi, P., He´thelyi, E´., 1975. Variations in the essential oils of Tanacetum vulgare. Phytochemistry 14, 1539–1544.
Weyerstahl, P., Marschall, H., Schroder, M., Wahlburg, H.C., Kaul, V.K., 1997. The sesquiterpene fraction ofArtemisia laciniataWilld. Flavour Fragrance J. 12, 315–325.
Wilkomirski, B., Kucharska, E., 1992. Triterpene chemotypes of some Polish populations ofTanacetum vulgare. Phytochemistry 31, 3915–3916.
