Characterization of Essential Oils of Xylopia aethiopica ( Dunal ) A . Rich for Afforestation of the Coastal Savanna at Pointe-Noire ( Congo-Brazzaville ) 1

The essential oil from fruits, leaves and stem bark of Xylopia aethiopica of Congo-Brazzaville was obtained by steam distillation and analyzed by CG on two columns with different polarities (polar and apolar) and by CG/SM. The essential oil from fruits was characterized by the presence of three constituents at levels of at least 10%. These were pinenes (alpha-+beta-) as major components (17%), 1, 8-cineole (13.3%) and sabinene (10%), all monoterpene hydrocarbons. The three most abundant oxygenated monoterpenes were trans-pinocarveol (8.2%), myrtenal (6.3%) and myrtenol (6.2%). The essential oil from leaves was characterized by the presence of pinenes (alpha-+beta-) as major components (39-60%). Sesquiterpenes came second, with caryophyllene the most abundant (6-18%). Oil from stem bark was made up of pinenes (27-57%), with beta-cubebene (11-14%) in second position and transpinocarveol (6%) and myrtenal (5%) jointly in third position.


INTRODUCTION
Essential oils are a Non-Wood Forest Products (NWFPs).This FAO term designates all biological materials that are exploited other than timber and other woody raw materials.NWFPs include a broad diversity of useful products, including foods, spices, medical drugs, forage, essential oils, resins, gums, latex, tannins, dyes, cane, fibres, bamboos and all sorts of ornamental animal and plant products (CEE-FAO, 1999).
Essential oils are well-placed to enter in official trade statistics, but the potential value of NWFPs is difficult to estimate accurately and is indeed often under-estimated because such products are mostly used locally.
Xylopia aethiopica is a dense forest understory tree 15-30 m high and 60-75 cm in diameter, growing on river banks or marshland.It has a slender trunk with a buttressed base 50 cm to 1 m in diameter (Thomas, 1969).It is relatively tall and topped by a plume of branches and twigs spread out horizontally.It belongs to the Annonaceae family, subfamily Annonoideae, tribe Unoneae, subtribe Xylopineae.This forest tree is found in the Congo Basin, where it has many uses: Its strongly peppery seeds and carpels are used as spices or condiments (Aubreville, 1959;Tchiegang and Mbougueng, 2005;Letouzey, 1982;Tisserant, 1950).Xylopia aethiopica is also used to treat scabies (Letouzey, 1982), asthma, stomach pains, rhumatism, malaria (Burkill, 1985;Fekam et al., 2003), cough, bronchitis, dysentery, female sterility and abdominal pains (Anvam, 1998).
In Congo, where it is exploited essentially for firewood, Xylopia aethiopica is under strong human pressure.To ensure its sustainability, commercial use is envisaged by the production of essential oils for food and medicine through a programme of afforestation of poor coastal savannas in the region of Pointe-Noire.This option has been taken by the Coastal Forest Research centre (CRFL), Congo-Brazzaville, with extremely encouraging results (Fig. 1).However, such an endeavour implies acquiring basic knowledge of the forestry, chemistry and technology involved.
The study described here concerns the quantitative and qualitative characterization of essential oils of Xylopia aethiopica extracted from the fruits, leaves and bark of naturally-growing trees.

MATERIALS AND METHODS
Plant material: The fruits (Fig. 2), leaves and bark of X. aethiopica studied were collected wild in the Youbi forest, 100 km north of the town of Pointe-Noire.The Youbi forest is located between latitudes 4°0'00'' and 4°30'00'' South and longitudes 11°30'00'' and 12°0'00''.Annual rainfall in the region is about 1400-1500 mm.The harsh dry season lasts from June to September and the rainy season from October to May.The samples were collected in September 2010.

Extraction of essential oils:
The essential oils were obtained by steam distillation.Water and plant material (300 g of dried fruits) were placed in a 500 mL roundbottomed flask and boiled for 4 h.The organic phase of the resulting condensate was separated from the aqueous phase by extraction with diethyl ether.The organic phase was dried over sodium sulphate and the essential oil was recovered after evaporation of the solvent.

Determination of the chemical composition of the essential oils:
After detailed analysis of the essential oil extracted from fruits (protocol 1), which identified more than 70 constituents, we undertook a systematic study of the essential oils from the different plant parts with a simplified experimental protocol (protocol 2).
Experimental protocol 1: GC analysis: GC analysis was carried out using a Perkin-Elmer Auto System apparatus (Waltham, MA, USA) equipped with a dual Flame Ionization Detection (FID) system and the fused-silica capillary columns: Rtx-1 (polydimethylsiloxane) and Rtx-wax (polyethyleneglycol).The oven temperature was programmed from 60 to 230°C at 2°C/min and then held isothermally at 230°C for 45 min.Injector and detector temperatures were maintained at 250 and 280°C.Samples were injected in the split mode (1/50) using helium as carrier gas (1 mL/min) and a 0.2 µL injection volume of pure oil.Retention Indices (RI) of compounds were determined relative to the retention times of a series of n-alkanes (C5-C30) (Restek, Lisses, France) with linear interpolation using the Van den Dool and Kratz (1963)  Experimental protocol 2: GC analysis: GC analyses was performed on a Hewlett-Packard 6890 equipped with a split/splitness injector (280°C, split ratio 1:10), using DB-5 column (30 m×0.25 mm, df: 0, 25 µm).The temperature program was 50°C (5 min) rising to 300°C at rate of 5°C/min.Injector and detector temperature was 280°.Helium was used as carrier gas at a flow rate 1 mL/min.The injection of the sample consisted of 1.0 µL of oil diluted to 10% v/v with acetone.
GC-MS analysis: GC/MS was performed on a Hewlett-Packard 5973/6890 system operating in EI mode (70 eV), equipped with a split/splitness injector (280°C, split ratio 1:20), using DB-5 column (30 m×0.25 mm, df: 0.25 µm).The temperature program was 50°C (5 min) rising to 300°C at rate of 5°C/min.Injector and detector temperature was 280°.Helium was used as carrier gas at a flow rate 1 mL/min.The identification was carried out by calculating retention indices and comparing mass spectra with those in data banks (Adams, 2001;Mc Lafferty and Stauffer, 1989).

Statistical treatment:
Means, standard deviations and the usual graphs were obtained with Excel software.

Essential oil contents in the different plant compartments:
In preliminary work, we evaluated the factors determining the quantitative and qualitative production of this oil: duration of the extraction and state of conservation of the raw material (storage conditions).This study was carried out on fruits.It is recognised that the fruit of X. aethiopica is the part richest in essential oil (Ayedoun et al., 1996;Karioti et al., 2004).It was therefore in this part that we sought the characteristic essential oil of the plant, assuming a significant intra-tree variability in its oil.On a reduced number of samples we then evaluated the levels of essential oil according to its storage compartments in the plant (fruits, leaves and bark).
Figure 3 shows the curve for the essential oil extraction from fruits as a function of extraction time during steam distillation.We found that after 100 min, the mass of essential oil extracted was 9.2 g.Extrapolation of the curve 1/m = f (1/t) to 1/t = 0 (t = infinity) gives 1/m = 104 g, corresponding to a total mass of 9.6 g of essential oil contained in the fruits studied, indicating an extraction rate of (9.2/9.6) 100 = 96% (Fig. 4).There was no need to prolong the extraction beyond 240 min.Accordingly we set our extraction time to 4 h for the rest of the study.
The second parameter evaluated was the influence of the conservation of the raw material on extraction yield.We see in Fig. 5 that fruits conserved at ambient temperature (30-35°C) lost one third of their extractable essential oil after 1 week and half after 2 weeks.By contrast, oil loss was negligible when the fruits were stored at 4°C for 2 weeks.
Like the fruits, leaves and bark also contain essential oils, but in smaller amounts.Table 1 gives the data for four trees taken at random in the wild forest.We can see that fruits, which serve as the main essential oil storage compartment in X. aethiopica, had levels of up to 7%.Leaves were the second richest storage compartment, with levels of 0.2-0.3% and bark contained only 0.02-0.07% of essential oil.
In a given compartment, we found very high amplitude variations in essential oil levels: 2-7% for fruits, 0.1-0.3% for leaves and 0.01-0.07%for bark.
There seems to be a correlation between essential oil content in fruits and that in leaves (Fig. 6), fitting the equation:      2) identified 71 constituents, representing nearly 92% of the total essential oil, with 31.4% monoterpene hydrocarbons.57.0% oxygenated monoterpenes, 3.2% sesquiterpenes and 1% diterpenes.
Of the 71 constituents of the essential oil, 16 present at levels of more than 1% totalled 81% and so can be considered as characteristic of the oil from the fruits studied.Three constituents were present at a levels of at least 10%: these were 1.8-cineole (13.3%) and sabinene (10.0%), pinenes (17.0%) in particular alpha-pinene (6%) and beta-pinene (11.5%), all monoterpenes hydrocarbons.
With only three constituents exceeding an individual content of 10%, we have an essential oil with a large number of constituents each present at rather low levels, giving a relatively complex radar plot (Fig. 6 and 7).
As an illustration, Table 4 gives the composition of essential oils extracted from fruits, leaves and bark limited to ten major components making up some 80% of the total composition of the essential oil extracted from fruits.
For a given tree, for example Tree 1, the essential oils extracted from the fruits, leaves and bark yield the following total radar plot (Fig. 8).
In sum, we find that X. aethiopica gives relatively complex essential oils, which vary in particular according to the plant compartment that produces them: fruits, leaves or bark.

Intra-tree variability of essential oils:
A comparative examination of the compositions of essential oils from the different plant compartments (Table 4) shows that pinene (alpha-and beta-) occurred in all these compartments as by far the most abundant component.Sabinene seems to be characteristic of the fruits, in which it occurred at fairly high levels (>20%) and was even the foremost major component in one of the four trees studied.It was entirely absent from leaves and bark.
Caryophyllene was more abundant in the leaves, either as the isomer beta-caryophyllene (5-9%), or in its oxidised form, caryophyllene oxide (5-6%).It was found in the bark of one tree among the four studied.
We note, however, that 1.8-cineole was almost absent from leaves except in Tree 11, where it was present in an appreciable amount (14%).Bark contained beta-cubebene at fairly high levels (11-14%) in two of the four trees.
Inter-tree variability of essential oils: The lowest inter-tree variability was found for essential oil extracted from fruits, which had a profile based on pinenes (alpha and beta) and sabinene (40-60%), one or the other predominating.The oxygenated monoterpenes, which were high in the total oil, came third for individual content, with myrtenal leading (6-7%).
These oils displayed a quasi stable composition from one tree to another, as illustrated in Fig. 9.
Starting with Tree 1 as a basic morphogram (radar plot), we can see that: Z-beta-ocimene appeared in Tree 5 with levels well below 10% and the pinene/sabinene ratio was inverted and 1.8-cineole appeared in Tree 11 (likewise for Tree 15).
The oils extracted from the leaves contained practically no sabinene.Although the chemotype remained pinene-rich (39-60%), sesquiterpenes came second, with caryophyllene leading (6-18%).Oxygenated monoterpenes were totally absent.The oils extracted from tree bark were also made up of pinenes (27-57%).It was in this compartment that we found the greatest variability of the second most abundant constituent; in decreasing order, we had: trans-pinocarveol, myrtenal and beta-cubebene (11-14%).
Thus for example, for the four trees studied, we had widely different profiles outside pinene: In sum, this preliminary study indicates that the content and composition of essential oils varied over a relatively broad scale of values, in particular when they were extracted from a naturally-growing population of trees, here the Youbi forest at Pointe-Noire, Congo-Brazzaville.
For the same tree compartment, e.g., the fruits, we are struck by the high variability in the composition of their oils according to the harvesting location across Africa (Table 5 and 6).

CONCLUSION
Essential oils extracted from wild forest trees varied qualitatively and quantitatively to a noteworthy degree: • Among the different plant storage compartments • From one tree to another The fruits were the main storage compartment for the essential oils, with levels of up to 7%.
The leaves contained on average 30 times less essential oil than the fruits and the bark 100 time less.
We note a linear correlation between the essential oil contents of the fruits and of the leaves.If this finding was confirmed by the study of a more representative sample, we could predict the oil content of young trees in nurseries by analyzing their leaves.
The oils in all the compartments were mostly made up of pinenes; sabinene seemed to be characteristic of fruits, beta-caryophyllene and its oxide of leaves and beta-cubebene of bark.
The results of this study give an overall picture of a complex essential oil composition, based on a constant monoterpene profile with pinenes as major constituents.
Ultimately, selection of individuals would be desirable, based on what essential oil content or major constituents are sought.

Fig. 1 :
Fig. 1: Xylopia aethiopica plantation in coastal savanna (Loandjili, Pointe Noire, Congo Brazzaville, 6 years old trees) equation and software from Perkin-Elmer.GC-MS analysis: Samples were analyzed with a Perkin-Elmer turbo mass detector (quadrupole) coupled to a Perkin-Elmer Auto System equipped with the fused-silica capillary columns Rtx-1 and Rtx-wax.Carrier gas: helium (1 mL/min), ion source temperature: 200°C, oven temperature programmed from 60 to 230°C at 2°C/min and then held isothermally at 230°C (35 min), injector temperature: 280°C, energy ionization: 70 eV, electron ionization mass spectra were acquired over the mass range 35-350 Da, Identification of individual components was based: • On comparison of calculated RI on polar and apolar columns, with those of authentic compounds or literature data • On computer matching with commercial mass spectral libraries and comparison of mass spectra with those of our own library of authentic compounds or literature data.Components relative concentrations were calculated based on GC peak areas without using correction factors (König et al., 2001; NIST (National Institute of Standards and Technology), 1999; Anonymous, 2005; Adams, 2001) Fig. 3: Variation of the mass of extracted essential oil vs. extraction time

Fig. 6 :
Fig. 6: Correlation between fruit essential oil vs. leaf essential oil yield (%) on a same tree

Fig. 8 :
Fig. 8: « Radar plot » representation of fruit essential oil composition from natural forest X. aethiopica in the different compartments on the same tree (A1)

Table 1 :
Essential oil yield (%) in different compartments of the tree S.D.: Standard deviation

Table 2 :
Chemical composition of essential oil extracted from the wild forest fruits of X. aethiopica RI: Retention index; apol: Apolar column; pol: Polar column

Table 4 :
Yield and composition of fruit leaf and bark essential oils from 4 wild forest trees A1

Table 5 :
Variation of fruit essential oil composition of Xylopia aethiopica vs. harvesting countries in Africa

Table 6 :
Variation of leaf essential oil composition of Xylopia aethiopica vs. harvesting countries in Africa