Essential Oils from Bolivia. III. Asteraceae: Artemisia copa Philippi

Journal of Essential Oil Research: JEOR, Nov/Dec 2004 by Arze, Javier Bernardo Lopez, Collin, Guy, Garneau, François-Xavier, Jean, France-Ida, Gagnon, Hélène

Abstract

The chemical composition of the essential oil obtained from Artemisia copa growing in the high valley region of Bolivia, Province of Cochabamba, was determined by a combination of GC and GC/MS measurements. The main constituents were β-thujone (42%), chamazulene (6.5%) and a homoterpene (6.0%) as well as limonene (5%) and α-pinene (4.8%).

Key Word Index

Artemisia copa, Asteraceae, essential oil composition, β-thujone, chamazulene, homoterpene.

Introduction

In the previous papers of the series, we initiated a systematic study of the Bolivian flora, particularly that occurring in the high valley region. The first papers dealt with Baccharis tricuneata varruiziana (1) and Ophryosporus heptanthus (2), two plants belonging to the Asteraceae family. In the present study, we present the results obtained from another member of the same family: Artemisia copa. This oil was studied many years ago by several researchers from Argentinean universities. Some of the physical properties were given as well as the composition of the oil from plants growing in northwest Argentina (3-5). This plant grows also in northern Chile, in regions close to Bolivia.

Experimental

Plant material: Artemisia copa Philippi. (local name: Ajenjo, Sandiobugazu) is a shrub 30 to 60 cm tall, highly branched and possessing fragrant, small and abundant oblong leaves with glandular points. Numerous capitulums, briefly petiolated, disposed in the under arm branch axila of the superior leaves forming untrue bunches of leaves. Yellowish flowers, dimorphs; the disc flowers are hermaphrodite. Cilindroid aquenus with 2 mm long.

This species was introduced and can be found in the mountain forests of the warm tropical valleys of the Oriental Andean area in Cochabamba, at 2300 m above sea level. It occurs in shallow soils, with high content of organic material and semi humid. It grows in plains and slightly inclined hills. They can be found surrounding cultivated lands and near farmhouses.

Two thousand g of plant material was collected in Rodeo Adentro in the tropical zone of the Carraco Province of Cochabamba and in the high valleys around La Paz (altitude: 3600 m above sea level). Voucher specimens have been deposited in the Herbarium Nacional Forestal Martin Cardenas (BOLV) de la Universidad Mayor de San Simon.

Oil extraction: The aerial parts of the plant material (2000 g) were introduced into the distillation pot with 6 L of water. Distillation for 1.5 h gave 7.5 mL of oil, producing a yield of 0.375% (volume/weight).

GC and GCMS: GC analyses were done on an HP 5890 gas Chromatograph equipped with a polar Supelcowax 10 column (30 m × 0.25 mm, film thickness 0.25 µm) and a nonpolar DB-5 column (30 m × 0.25 mm, film thickness 0.25 µm). The temperature program was 40°C for 2 min, then 2°C/ min to 210°C, and held constant for 33 min. The compounds were identified by their retention indices on both columns and by GC/MS. GC/MS analyses were done on an HP 5972 MSD mass spectrometer at 70 eV coupled with an HP 5890 gas Chromatograph (1,2). The GC/MS analyses were performed on both columns, polar and non-polar, with the same oven operating conditions. Retention indices and mass spectra were compared to those of the literature (6) and to our own databases (7).

Results and Discussion

Physical properties: The yield of oil was 0.375% (volume/weight). Its relative density was 0.9774 (20/20) and it was too dark for the determination of its optical rotation. The color of the oil was dark blue, with a pleasant odor. These properties are different from those previously reported: d(20/4) = 0.9147 and n^sub D^(20) = 1.4559 (3,4). As it will be seen later, these differences are the result of a different composition.

The main constituents of this oil were β-thujone (~42%), chamazulene (~6%) and, as discussed in the following section, a homoditerpene derivative (~6%). The other constituents belong to the monoterpene group (~14%) with limonene (~5%) and α-pinene (~4.8%) as the main compounds. Germacrene D, bicyclogerrnacrene and β-caryophyllene (~2-3% each) are the main components of the sesquiterpenes, the second important group (~5%). Finally, each of the oxygen-ated monoterpenes and sesquiterpenes represents 5% of the total, spathulenol and nerolidol (~1% each) being the most important members.

The oil of the present sample of A. copa was characterized by its blue color that can be attributed to the presence of chamazulene, although some of the chamazulene derivatives may also contribute to the visible light absorption. The GC analysis of this sample appears in Table I. Except where otherwise indicated, all products have been identified through the combination of retention indices on nonpolar and polar columns and the known mass spectrum (6,7). Five compounds were identified by comparison of their mass spectra with those published for the constituents found in the oil of Tanacetum annuum (8,9). As can be seen, the mass spectra are quite similar except in the case of 3,6-dihydrochamazulene, where relative peak intensities are slightly different. In order to ascertain the proposed identification, we have made the comparison of the reported retention times. First, Greche et al. gave the retention indices for the four chamazulenic compounds on the apolar DB-5 column (Table II). They are close to our observations except for 3,6-dihydrochamazulene (9). For the latter compound, the difference in the retention index was much greater than what was observed for the other compounds. This kind of comparison on the polar column is not as easy as just observed since only the retention times were published (8). Using our own retention indices for 10 compounds from α-pinene and camphene to chamazulene and the published retention times, one can calculate the corresponding retention indices. This method is only indicative since not only both chromatographs and their use are not rigorously identical, but above all, the oven temperature program is quite different: 2°C/min in our case against 5°C/ min in the reported paper (8). Again, one can see that the socalculated retention indices are in a relatively good agreement in both laboratories (Table II). This is particularly true for the 5,6-dihydrochamazulene and 7,12-dehydro-5,6,7,8-tetrahydrochamazulene. For the two homoditerpenes, their retention indices are relatively in good agreement although the order of elution is reversed. Finally, the calculated retention index for 3,6-dehydrochamazulene is smaller by ca 200 units from our own value. Thus, the identification of the so-called 3,6-dihydrochamazulene is probably doubtful.