Journal of Oleo Science
Copyright ©2010 by Japan Oil Chemists’ Society
J. Oleo Sci. 59, (7) 351-360 (2010)
Triterpene Alcohol and Fatty Acid Composition of
Shea Nuts from Seven African Countries
Toshihiro Akihisa1 ＊ , Nobuo Kojima1, Naoko Katoh1, Yuki Ichimura1, Hirohisa Suzuki1,
Makoto Fukatsu1, Steven Maranz2 and Eliot T. Masters3
1
College of Science and Technology, Nihon University (1-8 Kanda Surugadai, Chiyoda-ku, Tokyo 101-8308, JAPAN)
Department of Microbiology and Immunology, Weill Medical College, Cornell University (1300 York Ave, New York, NY, U.S.A.)
3
World Agroforestry Centre (ICRAF) (PO Box 30677-00100 Nairobi, KENYA)
2
Abstract: The content and composition of triterpene alcohol fractions of the non-saponifiable lipids (NSL)
along with the fatty acid composition of the kernel fats (n-hexane extracts) of the shea tree (Vitellaria paradoxa; Sapotaceae) were determined for 36 samples from seven sub-Saharan countries: Cote d Ivoire, Ghana, Nigeria, Cameroun, Chad, Sudan, and Uganda. The fat content of the kernels, proportion of NSL in the
fats, and triterpene alcohols in the NSL are in the range of 30–54, 2–12, and 22–72％, respectively. The triterpene alcohol fractions contained α-amyrin (1), β-amyrin (2), lupeol (3), and butyrospermol (4) as the major constituents along with minor or trace amounts of ψ-taraxasterol (5), taraxasterol (6), parkeol (7),
24-methylene-24-dihydroparkeol (8), 24-methylenecycloartanol (9), dammaradienol (10), and 24-methylenedammarenol (11). Fatty acid composition is dominated by stearic (28–56％) and oleic (34–61％) acids.
Shea butters from West African provenances contained in general higher levels of triterpene alcohols and
stearic acid than those from East African provenances. Both stearic acid and total triterpene alcohol contents were significantly correlated to the latitude and elevation of the source population, indicating that
higher levels of these compounds are found at higher ambient temperatures.
Key words: Vitellaria paradoxa, shea nuts, triterpene alcohol, fatty acid
1 INTRODUCTION
The shea tree［Vitellaria paradoxa C. F. Gaertn.; synonyms Butyrospermum paradoxum（C. F. Gaertn.）Hepper, Butyrospermum parkii（G. Don）Kotschy; family Sapotaceae］is indigenous to the savanna belt extending
across sub-Saharan Africa north of the equator, ranging
from Mali in the west to Ethiopia and Uganda in the east
（extending from 16°W to 34°E longitude and 1°N to 15°
（Fig. 1）. The most valued product of shea
N latitude）1-4）
tree is shea fat（shea butter）, the fat extracted from the
kernels. Processed shea fat is used primarily as a cocoa
butter additive in chocolate manufacture1）. These properties are due to the structure of its component triglycerides.
In addition, shea fat is increasingly popular in skin care
products and cosmetic product formulations, in part due to
the unusually high level of non-saponifiable lipid
（NSL）constituents in the fat5）. The main NSL constituents of shea fat
have been reported to be triterpene alcohols such as
α-amyrin（1）, β-amyrin（2）, lupeol（3）, and butyrospermol
（Fig. 2）. Since naturally occurring triterpene alco（4）6-9）
hols and their derivatives exhibit a variety of biological activities including anti-inflammatory, antitumor, chemopreventive, and antimycobacterial activities10-14）, we wished to
explore and determine whether significant differences in
qualitative and quantitative compositions exist for the NSL
of shea fat among widely dispersed V. paradoxa populations. We now report, in this paper, the contents and composition of the triterpene alcohol fractions from the NSL
and fatty acid composition of kernel fats obtained from 36
shea nut samples from seven sub-Saharan countries, and
observations on the relation between the triterpene alcohol
constituents and fatty acid composition.
2 EXPERIMENTAL
2.1 Materials and chemicals
Collection of shea nut samples was undertaken by one of
the authors（E.T.M.）using a standard collection protocol
＊
Correspondence to: Toshihiro Akihisa, College of Science and Technology, Nihon University, 1-8 Kanda Surugadai, Chiyoda-ku,
Tokyo 101-8308, JAPAN
E-mail: akihisa@chem.cst.nihon-u.ac.jp
Accepted January 13, 2010 (received for review November 11, 2009)
Journal of Oleo Science ISSN 1345-8957 print / ISSN 1347-3352 online
http://www.jstage.jst.go.jp/browse/jos/
351
T. Akihisa, N. Kojima, N. Katoh et al.
Table 1 Longitude, Latitude, and Elevation of the Sites of
Shea Nuts Collection
Country
Cote d’Ivoire
Latitude
Elevation
(m)
1
W 7º 12’ 58”
N 10º 1’ 35”
453
2
W 5º 6’ 37”
N 9º 10’ 57”
384
Ghana
3
W 0º 29’ 27”
N 6º 47’ 11”
145
Nigeria
4
E 4º 28’ 48”
N 11º 35’ 3”
280
5
E 5º 18’ 7”
N 10º 58’ 9”
337
6
E 5º 21’ 30”
N 9º 14’ 49”
197
7
E 6º 13’ 52”
N 10º 10’ 22”
347
8
E 6º 44’ 43”
N 9º 5’ 50”
242
9
E 6º 54’ 45”
N 8º 35’ 4”
173
10
E 6º 56’ 35”
N 10º 38’ 44”
683
11
E 6º 58’ 35”
N 10º 38’ 45”
456
12
E 7º 27’ 9”
N 9º 40’ 53”
365
13
E 8º 58’ 54”
N 8º 38’ 34”
160
14
E 9º 10’ 11”
N 7º 17’ 11”
192
17
E 11º 34’ 36”
N 9º 9’ 16”
313
18
E 12º 29’ 6”
N 9º 19’ 15”
245
19
E 12º 52’ 13”
N 9º 33’ 36”
267
15
E 10º 28’ 49”
N 5º 12’ 55”
1421
16
E 11º 2’ 56”
N 5º 50’ 12”
988
20
E 13º 15’ 42”
N 7º 14’ 57”
974
21
E 14º 19’ 34”
N 9º 36’ 30”
314
22
E 15º 29’ 30”
N 9º 35’ 54”
363
23
E 15º 38’ 17”
N 9º 19’ 41”
400
24
E 16º 9’ 17”
N 8º 31’ 25”
468
25
E 16º 21’ 12”
N 9º 22’ 45”
366
26
E 17º 4’ 44”
N 8º 38’ 16”
390
27
E 17º 26’ 45”
N 9º 1’ 12”
387
28
E 18º 2’ 19”
N 9º 3’ 56”
418
29
E 28º 26’ 55”
N 7º 17’ 19”
473
30
E 30º 31’ 27”
N 6º 34’ 57”
471
31
E 33º 11’ 58”
N 2º 27’ 10”
1207
32
E 33º 13’ 0”
N 3º 3’ 0”
1151
33
E 33º 15’ 7”
N 2º 19’ 19”
1203
34
E 33º 15’ 43”
N 2º 16’ 23”
1202
35
E 33º 19’ 27”
N 2º 40’ 59”
1213
36
E 33º 40’ 59”
N 2º 21’ 0”
1203
Fig. 1 Map of Africa Showing Sites (•) of Sampled
Shea Nuts in Relation to the Species Distribution
Range that Extends from 16°W to 34°E
Longitude and 1°N to 15°N Latitude.
Cameroun
Chad
Fig. 2 Structures of 11 Triterpene Alcohols Identified
in the Non-saponifiable Lipid Fraction of Shea
Kernel Fat.
developed by one of the authors（S.M.）under contract to
World Agroforestry Centre
（ICRAF）
, with subsequent laboratory treatment and sample preparation and transfer to
Nihon University supported by ICRAF in Samanko（Mali）,
Yaoundé, and Nairobi. Shea nut samples were collected
during the 2006 shea season（May through July）from
healthy mature trees at 36 sites of seven countries in Africa
（Table 1 and Fig. 1）
. These included two sites in Cote d Ivoire, one in Ghana, 14 in Nigeria, three in Cameroun,
eight in Chad, two in Sudan, and six in Uganda. The nut
samples were numbered sequentially based on the longitude of the collection site, beginning with the westernmost
site. Standard fatty acid methyl esters were purchased
from Sigma-Aldrich Japan Co.
（Tokyo）
.
Longitude
Sudan
Uganda
352
J. Oleo Sci. 59, (7) 351-360 (2010)
Triterpene Alcohol and Fatty Acid Compositions of Shea Fat
2.2 Kernel fat content
Whole nuts were oven-dried at 60℃ over 72 h and decorticated. Kernels were crushed in a mortar and finely
ground in a coffee mill. Pulverized samples were weighed
and extracted with n-hexane under reflux three times for 3 h
each. Fat percentage was determined gravimetrically by
dividing extracted total fat by kernel weight after evaporation of solvent.
2.3 Saponification and separation of triterpene alcohol
fraction
The n-hexane extract was subjected to alkaline hydrolysis
（5％ KOH in MeOH, reflux, 3 h）followed by diisopropyl
ether extraction, yielding a neutral NSL fraction. The percentage of NSL in the fat was determined gravimetrically
by dividing the weight of the extracted NSL by the n-hexane extract weight after evaporation of the solvents. The
NSL fraction was further fractionated on 20×20 cm plates
coated with a 0.5 mm layer of Silica gel G60
（Merck & Co.,
Inc.）
. The sample
（25 mg）was applied to the plate and developed with n-hexane–ethyl acetate
（EtOAc）
（8:2）for 1 h.
The plate was sprayed with 0.1％ rhodamine 6G solution in
MeOH and observed under UV light. The portion of the
TLC plate containing triterpene alcohols（Rf 0.55）was cut
off and quantitatively extracted with n-hexane–EtOAc
（7:3）. The percentage of triterpene alcohols in the NSL
was determined gravimetrically by dividing the extracted
triterpene alcohol fraction by the NSL fraction. The triterpene alcohol fraction was acetylated in acetic anhydride
and pyridine overnight at room temperature. The acetate
fraction was used for GLC analysis and for isolation of individual components by argentation TLC followed by preparative HPLC.
2.4 Isolation and identification of triterpene alcohols
Shea nut sample no. 12（from Nigeria）was selected for
in-depth analysis. The triterpene alcohol fraction（2704
mg）gave an acetate fraction of 2887 mg after acetylation.
The acetate fraction was subjected to preparative argenta（developing solvent: cyclohexane–EtOAc, 95:5）,
tion TLC6）
,B
giving five fractions: fractions A
（1152 mg; Rf value 0.90）
（168 mg; 0.81）, C（789 mg; 0.79）, D（47 mg; 0.73）, and E
（48 mg; 0.60）. A portion of fraction A（30 mg）was loaded
onto a preparative reversed-phase HPLC column（ TSK
ODS-120A, 25 cm×10 mm i.d.; Toso Co., Tokyo）
and eluted
with MeOH
（flow rate: 3.0 mL/min）
, yielding α-amyrin acetate［ 1a; 18.2 mg; retention time（ t R ）50.8 min］6）and
β-amyrin acetate（2a; 3.1 mg; tR 46.8 min）6）. A portion（77
mg）of fraction B, separated using preparative HPLC as
above, gave butyrospermol acetate（ 4a; 2.9 mg; t R 32.4
6）
9, 14）
, ψ-taraxasterol acetate
（5a; 2.8 mg; tR 55.6 min）
,
min）
9, 14）
and taraxasterol acetate
（6a; 1.4 mg; tR 46.8 min） . Preparative HPLC as above for a portion
（74 mg）
of fraction C
gave lupeol acetate
（3a; 10.0 mg; tR 30.7 min）6）and 4a（10.3
mg）
. Fraction D, likewise on preparative HPLC, yielded 3a
（7.1 mg）, parkeol acetate（ 7a; 1.6 mg; t R 47.2 min）6）,
24-methylene-24-dihydroparkeol acetate（8a; 2.1 mg; tR
49.2 min）9, 15）, and 24-methylenecycloartanol acetate（9a;
6）
. Finally, preparative HPLC of fraction
5.2 mg; tR 76.5 min）
E gave 8a
（3.8 mg）
, dammaradienol acetate
（10a; 3.1 mg; tR
28.4 min）7）, and 24-methylenedammarenol acetate（11a;
7）
. The structures of these compounds
10.3 mg; tR 30.0 min）
are shown in Fig. 2. Identification of 11 triterpene alcohols
was established for the acetyl derivatives, 1a–11a, by comparison of their 1H NMR and MS data with those of corresponding compounds cited in the literature6, 7, 9, 14, 15）.
2.5 Measurement of triterpene alcohol fractions
Analysis of the triterpene alcohol fractions, as the acetyl
derivatives, was performed by GLC using a Shimadzu GC17A series gas chromatograph equipped with a FID. The
chromatograph was fitted with a DB-17 fused-silica capillary column（30 m×0.31 mm i.d., 0.25 μm film thickness,
J&W Scientific, Inc., Folsom, CA, U.S.A.）
. The column was
mL/min; split
operated isothermically at 275℃ with N（64
2
ratio 33:1）as a carrier gas. Injection and detector temperatures were 300℃. The retention times established for the
reference triterpene alcohols isolated from sample no. 12
（Nigeria）, described above, were used to identify the triterpene peaks for all samples. Relative retention times
（RtR）
were calculated with reference to an internal standard,
cholesterol acetate
（tR＝17.5 min）. The compounds and RtR
values were as follows: 1a（1.91）, 2a（1.69）, 3a（2.00）, 4a
（1.64）
, 5a
（2.46）, 6a
（2.60）
, 7a
（1.77）, 8a
（1.95）, 9a（2.00）
,
10a
（1.67）, and 11a
（1.79）
. The relative percentage of each
triterpene alcohol was obtained by dividing the individual
peak area by the sum of all peak areas obtained for triterpene alcohols using an integrator.
2.6 Preparation and analysis of fatty acid methyl esters
A few drops of sulphuric acid（0.15 μL）were added to a
solution of the n-hexane extract
（50 mg）in MeOH
（50 mL）
,
and the mixture was refluxed for 3 h. The solvent was
evaporated on a rotary evaporator and the methyl esters
thus prepared were extracted with diethyl ether 3 times
after the addition of water. The combined ether extracts
were washed with water until neutral, dried with anhydrous sodium sulphate, filtered and taken down to dryness
using a rotary evaporator. Fatty acid methyl esters were
analyzed with a Shimadzu GC-2014 series gas chromatograph equipped with a FID. The chromatograph was fitted
with a fused-silica capillary column（50 m×0.25 mm i.d.,
0.25 μm film thickness）designed for FAME applications
（Varian, Inc., Palo Alto, CA, U.S.A.）. The column was operated at 220℃ with He as a carrier gas（flow rate 157 mL/
min; split ratio 150:1）
. Detector and injection temperatures
were 300℃. Identification of fatty acids was done by the
use of standards. The relative percentage of each fatty acid
353
J. Oleo Sci. 59, (7) 351-360 (2010)
T. Akihisa, N. Kojima, N. Katoh et al.
was obtained by dividing the peak area of the fatty acid by
the sum of all fatty acid peak areas using an integrator.
2.7 Data analysis
Data was analyzed statistically using XLSTAT 2009
（Addinsoft, New York, NY, U.S.A.）. Regression analysis and
Spearman correlations were used to examine data trends
and to evaluate the relationship between chemical composition and geographic parameters.
3 RESULTS AND DISCUSSION
3.1 Kernel fat content
As shown in Table 2, the fat content of shea kernels investigated ranged from 29.7（sample no. 15 from Cameroun）to 53.7％（sample no. 12 from Nigeria）with a mean
value of 42.4％. No striking regional difference in fat content was observed among the samples examined except
those from Cameroun, which showed low fat percentage
with a mean value of 32.5％. Low kernel fat content for
Cameroon shea is consistent with that reported in the literature2）.
3.2 NSL content of kernel fat
The NSL content of kernel fats ranged from 2.4％（nos.
34 and 36 from Uganda）
to 11.6％
（no. 16 from Cameroun）
2）
with a mean 6.2％（Table . Whereas considerable variation was observed in the NSL content within a country
（e.g., 4.3–10.9％ for the samples from Nigeria）
, shea samples from western area of the shea belt（Cote d Ivoire,
Ghana, Nigeria, and Cameroun）exhibited higher NSL content than those from eastern area
（Sudan and Uganda）.
mol（4）, ψ-taraxasterol（5）, and taraxasterol（6）. Some
compounds co-eluted in the same peak: RtR 1.69 was identified as β-amyrin
（2）
and dammaradienol
（10）
, of which the
former was estimated to be the predominant component
based on the isolation experiment done for shea sample no.
12（from Nigeria）, described above. In a similar manner,
the peak with RtR 1.77 was estimated to consist of parkeol
（7）and 24-methylenedammrenol（11）, while RtR 1.91 consisted of α-amyrin（1）and 24-methylene-24-dihydroparkeol
（3）
and 24-methylene（8）
, and RtR 2.00 consisted of lupeol
cycloartanol
（9）. Taking these into consideration, the acetylated triterpene alcohol fractions of the shea samples examined were shown to contain α-amyrin（31.3–41.4％）as
the predominant component followed by lupeol（17.5–
25.1％）, butyrospermol（14.9–26.3％）, and β-amyrin（8.2–
13.2％）, accompanied by minor or trace amounts of compounds 5, 6, 8, 9, 10, 11, and several unidentified
compounds. An example of the GLC chromatogram for the
triterpene alcohol fraction（analyzed as acetyl derivative）
obtained from shea nut sample no. 1（from Cote d Ivoire）
was shown in Fig. 3.
Both the total triterpene alcohol content and the relative
percent composition of individual triterpene alcohols
showed significant correlations with geographic variables
（Table 4）. Higher total triterpene levels are associated
with higher latitude and lower elevation, both of which reflect hotter temperatures. Among individual triterpene alcohols, butyrospermol（4）most closely tracked the hot
temperature trend. Both α-amyrin（ 1）and β-amyrin（ 2）
showed an inverse tendency, correlating significantly with
3.3 Triterpene alcohol content
Triterpene alcohol fractions in the NSL ranged from
22.4％（no. 35 from Uganda）to 71.7％（no. 12 from Nigeria）
with a mean 48.1％（Table 2）
. As has been observed in
the NSL content described above, the proportion of triterpene alcohols in the NSL also varied considerably within a
country（e.g., 29.8–71.7％ for the Nigerian samples）. Shea
samples from western area of shea belt（Cote d Ivoire,
Ghana, Nigeria, and Cameroun）had a tendency to contain
more triterpene alcohols in the NSL than those from eastern area
（Table 2）
. Shea samples from Chad, located in the
central area of shea belt, showed intermediate levels of triterpene alcohols compared to Sudan and Uganda to the
east and Nigeria, Ghana, and Cote d Ivoire to the west.
3.4 Triterpene alcohol composition
The composition of triterpene alcohol fractions
（as acetyl
derivatives）from the NSL of 36 shea samples are shown in
Table 3. Among the eleven GLC peaks observed, RtR 1.64,
2.46, and 2.60 were respectively identified as butyrosper-
Fig. 3 GLC Chromatogram of the Triterpene Alcohol
Fraction (analyzed as the acetyl derivative)
Obtained from Shea Nut Sample No. 1 [see
footnote b) of Table 3 for the names of
compounds (1)〜(11)].
354
J. Oleo Sci. 59, (7) 351-360 (2010)
Triterpene Alcohol and Fatty Acid Compositions of Shea Fat
Table 2 Content of Shea Kernel Fats, Content of NSL in the Fat, and Contents of Triterpenes in the
NSL and in the Fat
Fat content of
shea kernel (％)
Country
Cote d’Ivoire
Ghana
Nigeria
Cameroun
Chad
Sudan
Uganda
Triterpenes in the Triterpenes in the
NSL (％)
fat (％)
1
46.1
8.2
57.5
4.7
2
45.0
7.6
47.3
3.6
SDa)
4.0
0.8
4.5
−
3
36.8
9.7
59.8
5.8
4
45.7
5.2
49.9
2.6
5
41.8
6.4
55.9
3.6
6
43.8
4.5
60.0
2.7
7
36.6
4.3
56.7
2.4
8
30.6
5.7
63.5
3.6
9
51.2
10.4
59.5
6.2
10
37.7
6.4
49.7
3.2
11
49.1
7.2
29.8
2.1
12
53.7
7.1
71.7
5.1
13
38.5
5.3
61.9
3.3
14
42.4
10.9
52.5
5.7
17
39.1
8.5
54.5
4.6
18
43.8
6.7
51.8
3.5
19
39.1
8.0
69.0
5.5
15
29.7
7.2
31.4
2.3
16
30.4
11.6
53.0
6.1
20
37.4
4.7
64.2
3.0
21
39.1
5.7
59.3
3.4
22
39.0
7.9
37.1
2.9
23
47.5
5.2
64.2
3.3
24
46.4
5.3
59.2
3.1
25
46.4
4.4
47.5
2.1
26
36.6
7.7
28.3
2.2
27
48.0
5.9
48.0
2.8
28
48.0
6.3
42.1
2.7
29
45.0
5.1
32.1
1.6
30
48.4
5.5
26.3
1.4
31
33.9
2.7
30.6
0.8
32
38.8
3.9
27.7
1.1
33
43.8
4.3
28.5
1.2
34
51.5
2.4
38.7
0.9
35
48.3
4.6
22.4
1.0
36
Mean (SD)
NSL in the fat
(％)
46.7
2.4
40.1
1.0
42.4 (6.1)
6.2 (2.2)
48.1 (13.7)
3.1 (1.5)
a) Standard deviations (n=3) for the content of shea kernel fat, content of NSL in the fat, and content of triterpenes in the NSL determined for sample no. 2.
355
J. Oleo Sci. 59, (7) 351-360 (2010)
T. Akihisa, N. Kojima, N. Katoh et al.
Table 3 Composition of Triterpene Alcohol Fractions from the Non-saponifiable Lipid Fractions of the Kernel Fats of
36 Shea Nut Samples Determined by GLC
RtRa) of individual
componentsb)
Shea nut sample
Cote d’Ivoire
Ghana
Nigeria
Cameroun
Chad
Sudan
Uganda
1
2
SDc)
3
4
5
6
7
8
9
10
11
12
13
14
17
18
19
15
16
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
1.54
1.64
(4)
1.69
(2) >> (10)
1.77
(7) > (11)
1.0
0.6
0.1
0.8
1.0
1.6
1.2
1.3
0.8
1.8
1.2
0.8
1.3
1.5
1.3
0.9
1.1
1.1
0.5
0.6
0.2
1.0
1.6
1.4
0.2
0.6
0.7
1.2
0.9
1.2
1.1
0.7
0.8
4.4
1.0
1.0
0.8
24.2
19.4
0.3
21.3
20.5
23.4
22.1
21.3
21.7
25.5
21.3
16.6
24.2
21.0
22.0
20.3
21.7
22.7
17.1
19.4
18.4
22.9
22.3
23.6
15.0
16.1
17.6
21.2
21.2
26.3
15.3
18.7
16.3
14.9
21.9
16.9
18.1
9.6
9.5
0.3
10.4
9.0
9.4
9.1
9.2
9.5
9.0
9.3
9.1
9.5
9.1
8.4
8.5
9.1
8.2
9.4
9.4
13.2
9.1
10.1
9.0
10.7
9.3
9.4
9.9
9.4
9.6
10.4
10.0
10.7
9.0
10.6
10.5
10.6
3.9
4.7
0.3
3.0
3.7
4.5
3.8
3.6
3.6
3.8
4.2
3.9
4.2
4.3
4.0
4.0
3.5
4.1
4.3
4.7
3.5
4.0
3.9
3.7
3.8
3.1
3.8
3.6
3.6
3.7
3.4
3.8
3.8
6.2
3.3
4.5
3.6
％ Composition
1.81 1.88
1.91
(1) >> (8)
0.1
0.6
0.1
0.5
0.2
0.4
0.2
0.2
0.5
0.2
0.5
0.4
0.4
1.1
0.6
0.1
0.2
0.1
0.9
0.6
4.1
1.1
0.5
1.5
1.5
0.3
1.5
1.3
1.3
1.6
1.2
1.7
1.4
1.3
0.8
1.8
2.1
1.3
1.1
0.9
1.8
0.9
1.1
1.0
1.8
1.5
1.6
0.9
1.3
3.0
1.3
1.8
2.1
1.2
1.5
1.1
0.2
2.0
1.3
1.6
33.3
36.8
1.1
39.0
37.0
35.4
36.7
36.2
35.6
32.2
36.1
39.6
33.7
35.8
36.8
34.7
35.6
33.0
37.9
37.3
39.8
35.1
32.7
33.2
41.4
38.3
36.7
36.6
36.2
34.4
37.8
37.5
41.4
31.3
36.9
36.5
39.5
2.00
(3) >> (9)
2.03
2.46
(5)
2.60
(6)
22.3
20.8
0.5
19.9
21.1
18.1
18.7
21.1
21.3
21.3
20.5
22.8
22.0
19.8
19.6
24.1
22.1
22.7
23.5
21.1
19.1
20.6
21.3
21.9
22.6
24.9
21.2
19.7
20.4
17.4
22.9
21.1
19.5
25.1
18.5
20.1
18.6
0.6
1.6
0.3
0.9
1.0
1.2
1.4
1.2
1.1
0.9
1.4
1.0
1.0
1.2
1.5
1.2
0.9
1.2
1.1
1.3
1.3
0.7
1.1
0.7
0.9
1.3
1.8
1.4
1.4
1.2
1.4
1.5
1.8
0.9
1.3
1.7
1.9
2.5
2.9
0.4
2.3
3.2
3.3
3.6
3.0
3.1
2.8
3.0
3.3
1.4
3.0
3.1
3.2
3.3
3.5
2.7
2.9
2.4
3.1
3.3
3.1
3.2
3.2
3.8
3.2
3.3
2.9
3.6
3.2
3.0
2.3
2.9
4.2
3.1
1.0
1.6
0.3
0.9
1.7
1.6
1.8
1.5
1.4
1.3
1.5
1.6
0.7
1.7
1.6
1.6
1.8
1.7
1.5
1.6
1.1
1.7
2,2
1.8
1.2
1.7
1.9
1.9
1.8
1.2
2.0
1.4
1.6
1.6
1.6
2.2
1.7
a) RtR=Relative retention times for the acetyl derivatives. Retention time for cholesterol acetate (17.5 min) is taken as 1.00.
b) α-Amyrin (1), β-amyrin (2), lupeol (3), butyrospermol (4), ψ-taraxasterol (5), taraxasterol (6), parkeol (7), 24-methylene-24dihydroparkeol (8), 24-methylenecycloartanol (9), dammaradienol (10), and 24-methylenedammarenol (11).
c) Standard deviations (n=3) deteremined for sample no. 2.
356
J. Oleo Sci. 59, (7) 351-360 (2010)
Triterpene Alcohol and Fatty Acid Compositions of Shea Fat
Table 4 Spearman Correlations (rs) between Geographic Parameters and the Dominant Fatty Acids, Total Triterpene
Alcohol Content, and the Four Major Triterpenes in Shea Fat, based on Analyses of 36 Shea Nut Samples
Stearic acid
(18:0)
Variables
rs
p-value
Oleic acid
(18:1)
rs
p-value
Triterpenes in fat
(％)
rs
0.392
p-value
α-Amyrin (1)a)
％
rs
p-value
Lupeol (3)b)
％
rs
p-value
Butyrospermol (4)
％
rs
p-value
β-Amyrin (2)c)
％
rs
p-value
Latitude
0.722 < 0.001
-0.680 < 0.001
0.018
-0.360
0.031
0.197
0.248
0.439
0.008
-0.453
0.006
Longitude
-0.758 < 0.001
0.748 < 0.001
-0.682 < 0.001
0.152
0.374
-0.065
0.705
-0.408
0.014
0.368
0.028
Elevation
-0.696 < 0.001
0.738 < 0.001
-0.663 < 0.001
0.337
0.045
0.002
0.992
-0.509
0.002
0.522
0.001
a) Eluted with 24-methylene-24-dihydroparkeol (8).
b) Eluted with 24-methylenecycloartanol (9).
c) Eluted with dammaradienol (10).
higher elevation and lower latitude, indicating increasing
relative abundance with cooler temperatures in comparison to other triterpenes. It should be noted that the absolute quantities of α-amyrin（1）and β-amyrin（2）are still
higher in lowland shea populations, due to the substantially
higher total triterpene alcohol content. Lupeol（3）had no
significant association with any of the parameters evaluated.
3.5 Fatty acid composition
Eighteen saturated and unsaturated fatty acids were
characterized in the shea kernel fat in this study
（Table 5）:
myristic（myr, 14:0）, pentadecanoic（pen, 15:0）, palmitic
（pam, 16:0）
, palmitoleic
（Δpam, 16:1, n–7）, heptadecanoic
（ept, 17:0）, stearic（ste, 18:0）, elaidic（ela, 18:1, trans-9）,
oleic（ ole, 18:1, n–9）, cis-vaccenic（ vac, 18:1, cis-11）,
linolelaidic（lnl, 18:2, trans-9, trans-12）, linoleic（lin, 18:2,
n–6）, linolenic（ lnn, 18:3, n–3）, arachidic（ ach, 20:0）,
eicosenoic（eic, 20:1, n–9）, behenic（ben, 22:0）, brassidic
（bra, 22:1, trans-13）
, erucic
（eru, 22:1, n–9）
, and lignoceric
（lig, 24:0）acids. Among these, the predominant fatty acid
was oleic acid（34.0–60.7％）followed by stearic（ 27.7–
55.7％）, linoleic（4.4–7.9％）, palmitic（1.9–5.0％）, and arachidic（tr–1.8％）acids. Stearic and oleic acids together accounted for 85–90％ of the fatty acids in most of the
samples. This fatty acid profile range is consistent with
previous studies reporting values for 432 shea samples in
42 populations in 10 African countries2）and for 158 shea
samples in 20 populations in four countries4）.
The fatty acid profiles of the shea samples analyzed show
a strong geographic trend, paralleling what we found with
the triterpene alcohols. Stearic acid content has a significant positive correlation with higher latitude, and a significant negative correlation with both increasing elevation
and easterly longitude（Table 4）. Oleic acid shows the inverse relationship. Since these two fatty acids are the dominant components of shea fat and are represented by relative percentage data, an increase in one is expected to be
accompanied by a decrease in the other. However, it appears that it is climate that primarily determines the bal-
Fig. 4 Correlation between the Dominant Fatty Acids
and Total Triterpene Alcohol Content of Extracted
Fat in 36 Shea Nut Samples.
a) Stearic acid is positively correlated (R2=0.29,
p=0.001) with triterpene alcohols, while b) oleic
acid is negatively correlated (R2=0.30, p < 0.001)
357
J. Oleo Sci. 59, (7) 351-360 (2010)
T. Akihisa, N. Kojima, N. Katoh et al.
Table 5 Fatty Acid Composition (％) of the n-Hexane Extracts of 36 Samples of Shea Nutsa)
％ Composition
myr
RtRb) of individual
components
pen
pam Δpam
ept
ste
ela
ole
vac
lnl
lin
lnn
ach
eic
beh
bra
eru
lig
14:0 15:0 16:0
16:1
17:0 18:0 18:1 18:1 18:1 18:2 18:2 18:3 20:0 20:1 22:0 22:1 22:1 24:0
0.79
1.00
1.03
1.09
1.18
3.4
tr
0.1
0.90
1.19
1.20
1.21
41.7
46.3
0.5
1.22
1.24
1.30
1.33
1.35
1.47
1.49
1.50
1.65
5.4
0.1
1.4
0.3
0.1
tr
tr
0.4
Shea nut sample
Cote d’Ivoire
1
tr
2
tr
3.3
tr
0.1
40.7
47.5
0.5
6.0
0.2
1.3
0.3
0.1
tr
0.1
SDc)
0.0
0.0
0.0
0.0
0.1
0.4
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.1
39.0
48.9
0.4
6.3
0.2
1.4
0.3
tr
Ghana
3
0.1
Nigeria
4
tr
5
tr
tr
6
7
Sudan
Uganda
tr
0.1
44.1
44.2
0.3
5.9
0.1
1.5
0.3
0.1
tr
tr
0.2
0.1
55.7
34.0
0.6
4.4
0.1
1.7
0.2
0.2
tr
tr
0.1
tr
0.1
45.7
0.5
43.7
4.9
1.6
0.2
0.2
40.6
0.4
48.2
5.6
0.2
1.3
0.2
0.1
7.3
0.2
1.4
4.6
42.8
0.5
43.0
9
4.3
40.0
0.2
44.5
0.9
6.8
tr
11
tr
tr
0.1
0.1
tr
0.1
44.7
43.8
0.6
5.2
0.1
1.4
0.3
0.2
0.1
0.2
tr
0.1
42.9
45.5
0.7
5.7
0.1
1.3
0.3
0.1
tr
0.1
3.6
tr
0.1
48.1
0.5
0.3
5.6
0.2
1.2
0.4
44.5
6.1
0.2
1.5
14
3.5
44.8
0.5
43.4
6.4
17
4.6
42.8
0.5
18
5.0
42.0
3.8
tr
16
1.9
tr
20
4.9
tr
tr
tr
23
tr
tr
24
tr
25
tr
26
tr
27
tr
28
tr
40.6
43.5
4.8
22
0.3
0.1
0.6
0.1
0.8
0.2
tr
0.1
0.2
1.0
0.2
tr
tr
6.3
0.2
1.2
0.3
0.1
tr
tr
5.1
0.1
1.2
0.3
0.1
tr
0.2
0.1
3.9
tr
0.1
39.3
48.8
6.1
0.1
1.2
0.2
0.1
tr
0.1
3.2
tr
37.8
52.2
5.0
0.1
1.1
0.3
0.1
tr
0.1
3.9
tr
39.3
49.1
5.9
0.1
1.2
0.2
0.1
tr
0.1
3.8
tr
33.6
54.4
5.9
0.1
1.2
0.3
0.1
tr
0.4
4.0
tr
41.0
47.4
5.3
0.1
1.3
0.2
0.1
tr
0.3
0.1
0.1
28.1
3.8
tr
0.1
3.7
tr
0.1
tr
tr
5.8
tr
0.1
tr
1.5
0.1
0.1
tr
7.9
1.0
tr
36
0.2
4.9
tr
0.5
51.1
tr
tr
6.3
0.1
0.1
0.7
48.9
3.9
tr
1.3
53.8
36.6
3.8
tr
1.4
0.1
51.5
39.3
31
35
0.2
5.7
0.1
30
34
7.3
tr
tr
tr
1.4
0.1
0.1
tr
0.7
0.1
4.1
4.7
tr
44.4
50.0
0.1
0.3
0.1
0.1
3.7
4.1
33
0.4
36.8
33.2
43.0
45.5
38.4
29
32
0.6
3.0
43.9
21
0.1
0.6
3.3
3.4
tr
0.1
0.1
1.8
13
15
0.1
tr
8
10
tr
3.0
3.5
19
Chad
0.1
2.9
3.3
tr
12
Cameroun
3.4
36.6
0.1
0.1
0.2
51.1
33.2
0.2
52.3
0.7
37.3
tr
50.3
0.6
59.1
0.8
31.2
56.5
0.9
27.7
60.7
tr
tr
5.9
0.1
1.0
0.1
7.5
0.2
1.4
0.3
0.4
5.7
0.1
1.4
0.3
0.1
tr
tr
6.3
0.2
1.0
0.3
0.1
tr
tr
5.8
0.1
1.0
0.2
0.1
tr
0.1
6.0
0.2
1.1
0.3
0.1
tr
tr
0.1
tr
3.8
0.1
0.1
30.1
58.2
0.7
5.0
0.1
1.0
0.3
0.1
3.9
tr
0.1
34.1
52.1
0.9
6.9
0.2
1.2
0.2
0.1
tr
0.2
3.8
tr
0.1
30.3
56.6
0.7
6.8
0.2
1.0
0.2
0.1
tr
0.1
a) Percentage composition of fatty acid determined by GLC as the methyl ester derivative. tr=trace amounts.
b) RtR=Relative retention times for the methyl ester derivatives. Retention time for methyl palmitate (45.4 min) is taken as 1.00.
c) Standard deviations (n=3) detemined for the fatty acid composition for sample no. 2.
358
J. Oleo Sci. 59, (7) 351-360 (2010)
0.5
Triterpene Alcohol and Fatty Acid Compositions of Shea Fat
ance between the two major fatty acids, since the temperature gradient across the shea savanna belt increases with
north latitude and decreases with elevation. The data from
this study show that the stearic acid and triterpene alcohol
contents of shea fat generally increase together, while oleic
acid content correlates negatively with total triterpene alcohols（Fig. 4）, again reflecting the underlying link with
temperature.
The data analysis also shows a significant longitudinal
trend
（Table 4）. This could be interpreted as reflecting genetic differences in keeping with the current taxonomic division into two subspecies, V. paradoxa ssp. paradoxa in
the West and V. paradoxa ssp. nilotica in the East2, 4）.
Among the shea nut samples investigated, those（nos. 29–
36）from ssp. nilotica, collected in Sudan and Uganda, are
compositionally distinct from the West African types. The
former possess consistently high oleic acid（50.3–60.7％）
and low（27.7–37.3％）stearic acid fat. However, samples
from Sudan, Chad, and Cameroun show intermediate values that indicate a continuum across the shea belt, rather
than two distinct populations
（Table 5）. A study of genetic
markers in shea populations in eight countries covering
most of the Vitellaria range showed that genetic distances
between populations were strongly correlated to geographic distances, showing an east-west incremental transition16）. The longitudinal effect seen in our results could reflect this gene flow; however, it seems clear from the strong
effect of elevation on shea lipid composition that temperature plays a major determinant role.
4 CONCLUSION
In this study, the kernel fats（n-hexane extracts）of 36
shea nut samples from seven sub-Saharan countries were
analyzed for the triterpene alcohol constituents in the nonsaponifiable lipid fractions and for fatty acid constituents.
Consistent with earlier findings, the kernel fats contained
unusually high levels of triterpene alcohols, ca. 0.8–6.2％,
although the values we measured were not as high as reported previously by di Vicenzo et al.4）. Four compounds,
α-amyrin（1）, β-amyrin（2）, lupeol（3）, and butyrospermol
（4）, constituted the major triterpene alcohols for all the
nut samples investigated. Fatty acid composition is dominated by oleic（34–61％）and stearic（28–56％）acids. The
triterpene alcohol content of shea nuts generally increased
with increasing stearic acid percentage, while decreasing
with greater oleic acid content. This trend reflects an association of shea lipid composition with climate, underscored
by significant correlations with both latitude and elevation.
Shea nuts from Ugandan provenances had substantially
lower triterpene alcohol content and higher oleic acid percentages than those from West Africa, with intermediate
values seen in central African provenances. The results of
this study will be of value for further utilization of shea kernel fat
（shea butter）
from specific origins according to product application in the cosmetic and pharmaceutical fields
in the future.
ACKNOWLEDGMENTS
Sampling of Vitellaria paradoxa across Africa in support of this study was implemented during 2006 with funding from the World Agroforestry Centre（ICRAF）with additional support from the Common Fund for Commodities
（CFC）from 2004 to 2007 under project CFC/FIGOOF/23
（Improving Product Quality and Market Access for Shea
Butter Originating from Sub Saharan Africa）. Sampling in
Ghana, Nigeria, Cameroon, Sudan, and Uganda was undertaken with the support of Bréhima Kone and Momoudou
Dia of ICRAF, by Nafan Diarrassouba
（in Côte d Ivoire）, by
Victor Keouna and Antcha Mady, and Aliouda Limane Worgué（in Chad）, in collaboration with the Bup Nde Divine
University of Ngaoundéré（in Cameroun）and Gordon Wagner of Lulu Works in Sudan.
References
1．Masters, E.T.; Yidana, J.A.; Lovett, P.N. Reinforcing
sound management through trade: shea tree products
.
in Africa. Unasylva No. 219, 55, 46–52
（2004）
2．Maranz, S.; Wiesman, Z.; Bisgaard, J.; Bianchi, G.
Germplasm resources of Vitellaria paradoxa based
on variations in fat composition across the species distribution range. Agrofor. Sys. 60, 61–69
（2004）.
3．Maranz, S.; Kpikpi, W.; Wiesman, Z.; de Saint Sauveur,
A.; Chapagain, B. Nutritional values and indigenous
preferences for shea fruits（ Vitellaria paradoxa
Gaertn. F.）in African agroforestry parklands. Econ.
Bot. 58, 588–600（2004）
.
4．di Vincenzo, D.; Maranz, S.; Serraiocco, A.; Vito, R.;
Wiesman, Z.; Bianchi, G. Regional variation in shea
butter lipid and triterpene composition in four African
countries. J. Agric. Food Chem. 53, 7473–7479
（2005）.
5．Alander, J. Shea butter – a multifunctional ingredient
for food and cosmetics. Lipid Technol. 16, 202–205
（2004）.
6．Itoh, T.; Tamura, T.; Matsumoto, T. Sterols, methylsterols, and triterpene alcohols in three Theaceae and
some other vegetable oils. Lipids 9, 173–184（1974）.
7．Itoh, T.; Tamura, T.; Matsumoto, T. 24-Methylenedammarenol: a new triterpene alcohol from shea butter.
Lipids 10, 808–813
（1975）.
8．Peers, K.E. The non-glyceride saponifiables of shea
butter. J. Sci. Fd. Agric. 28, 1000–1009（1977）.
359
J. Oleo Sci. 59, (7) 351-360 (2010)
T. Akihisa, N. Kojima, N. Katoh et al.
9．Itoh, T.; Uetsuki, T.; Tamura, T.; Matsumoto, T. Characterization of tritierpene alcohols of seed oils from
some species of Theaceae, Phytolaccaceae and Sapotaceae. Lipids 15, 407–411
（1980）.
10．Akihisa, T.; Yasukawa, K.; Kimura, Y.; Takase, S.; Yamanouchi, S.; Tamura, T. Triterpene alcohols from camellia and sasanqua oils and their anti-inflammatory effects. Chem. Pharm. Bull. 45, 2016–2023
（1997）
.
11．Fernández, M.A.; de las Heras, B.; García, M.D.; Sáenz,
M.T.; Villar, A. New insights into the mechanism of action of the anti-inflammatory triterpene lupeol. J.
Pharm. Pharmacol. 53, 1533–1539
（2001）
.
12．Akihisa, T.; Yasukawa, K. Anti-inflammatory and antiallergic properties of triterpenoids from plants. in Biomaterials from Aquatic and Terrestrial Organisms
（Fingerman, M.; Nagabhushanam R. ed.）, Science
Publ. Enfield. pp. 63–114
（2006）.
13．Akihisa, T. Anti-inflammatory, antitumor, and chemopreventive effects of triterpenes from plants and fungi.
Oleoscience 7, 445–453（2007）
.
14．Akihisa, T.; Yasukawa, K.; Oinuma, H.; Kasahara, Y.; Yamanouchi, S.; Takido, M.; Kumaki, K.; Tamura, T. Triterpene alcohols from the flowers of Compositae and
their anti-inflammatory effects. Phytochem. 43, 1255–
1260（1996）
.
15．Itoh, T.; Tamura, T.; Matsumoto, T. 24-Methylenelanost-9（ 11）- en-3β-ol, new triterpene alcohol from
shea butter. Lipids 10, 454–460（1975）.
16．Fontaine, C.; Lovett, P.N.; Sanou, H.; Maley, J.; Bouvet,
J-M. Genetic diversity of the shea tree
（Vitellaria paradoxa C.F. Gaertn）, detected by RAPD and chloroplast microsatellite markers. Heredity 93, 639–648
（2004）.
360
J. Oleo Sci. 59, (7) 351-360 (2010)