Food Chemistry 118 (2010) 357–363

Contents lists available at ScienceDirect

Food Chemistry
journal homepage: www.elsevier.com/locate/foodchem

Effect of farm and industrial processing on the amino acid profile of cocoa beans
E.I. Adeyeye *, R.O. Akinyeye, I. Ogunlade, O. Olaofe, J.O. Boluwade
Department of Chemistry, University of Ado Ekiti, P.M.B. 5363, Ado Ekiti, Nigeria

a r t i c l e i n f o a b s t r a c t

Article history: An investigation into the amino acid profiles of unfermented and fermented cocoa nibs, as well as pro-
Received 19 July 2008 cess-line cocoa nibs (P-LCN) and processed cocoa cake samples (PCCS) was carried out. In the unfer-
Received in revised form 26 March 2009 mented cocoa nibs, Glu (128 mg/g crude protein i.e. 128 mg/gcp) was the most abundant amino acid
Accepted 30 April 2009
whilst the most concentrated essential amino acid in the same sample was Leu (72.2 mg/g crude pro-
tein); in the fermented cocoa nibs, a similar trend was observed, with respective values of Glu
(153 mg/gcp) and Leu (62.4 mg/gcp). Lys (181 mg/gcp) was most abundant amino acid in P-LCN and
Keywords:
Ile (63.3 mg/gcp) was the second most abundant; also, in PCCS, Lys (52.7 mg/gcp) was the most abundant
Cocoa beans
Farm fermentation
amino acid but Asp (43.7 mg/gcp) was the second most abundant. The total amino acid content was (mg/
Industrial processing gcp), 641 (unfermented nibs), 708 (fermented nibs), 635 (P-LCN) and 368 (PCCS), with corresponding
Amino acids essential amino acids of 300, 287, 478 and 185 (all with His), respectively. Based on whole hen’s egg,
the limiting amino acids for the samples were: Ser (unfermented nibs), Met (fermented nibs), Ala
(P-LCN) and Val (PCCS), whereas under provisional amino acid scoring pattern, they were: Met + Cys
(unfermented nibs), Met + Cys (fermented nibs), Thr (P-LCN) and Val (PCCS). Prolonged and high heat
treatments appeared to have reduced the essential amino acids of the PCCS as compared to the P-LCN.
Significant differences existed between contents of essential amino acids and non-essential amino acids
at p < 0.05 in unfermented cocoa nibs, fermented cocoa nibs and P-LCN.
Ó 2009 Elsevier Ltd. All rights reserved.

1. Introduction origin, hence the name – ‘‘Theobroma” meaning, food for the gods.
The drink has aphrodisiac properties and was held in high esteem
Cocoa bean was the foremost Nigerian foreign exchange earner as a nuptial aid.
before the advent of crude oil and it is currently in the second po- To obtain cocoa, the harvested pods are fermented, by naturally
sition after petroleum. The annual production in Nigeria is about occurring bacteria and yeasts to eliminate their natural, bitter,
165,000 metric tonnes (MT) (Akinyeye, 1999). A small percentage astringent quality, during which the seeds are cured and roasted.
of the annual production is processed, locally, to cocoa butter The clean kernels obtained after the removal of the shell, called
and cocoa cake, whilst the bulk is exported. The cocoa beans of cocoa nibs, are manufactured into various products. The larger
commerce are the seeds of the tree –Theobroma cacao (Linnaeus), percentage of the nibs is fat, removed by pressure, and is called
properly harvested, fermented and dried. It originated from Latin cocoa butter which is used in fine soaps and cosmetics and in med-
America about 500 years ago and thence to Europe, from where icine for emollients and suppositories. The residue is ground to a
it was introduced to other parts of the world (International Cocoa powder (cocoa) and used for beverages and flavouring. Chocolate
and Commodities Organisation, 2000). Cocoa bean is an oil seed, is a product in which the cocoa butter has been retained. Cocoa
just like palm kernel, groundnut, sesame seed or any of the other products have a high food value because of the large proportion
oilseeds. However, it is rarely processed in the same way as the of fat, carbohydrates, and protein. Cacao is classified in the division
other oil seeds in order to get the oil. The reason for this is probably Magnoliophyta, class Magnoliopsida, order Malvales and family
due to the unique physicochemical characteristics of the beans and Sterculiaceae.
its constituents, especially the fat. The Aztecs (of Mexico) prepared Cocoa products are eaten mainly because they are liked, by
the first cocoa drink called ‘‘Chocolatl” about 500 years ago (Minife, young and old, owing to their attractive flavours and appearance
1989). This chocolate drink was prepared from a mixture of which give pleasure in eating (Minife, 1989). The nutritional
ground, roasted whole beans or ‘‘nibs” and sugar. The drink was ex- parameters of cocoa are determined largely by the chemical com-
tremely rich because of the high fat content. The richness of the position of the material. The energy contribution to daily diets is
beverage made the Aztecs believe that the cocoa tree was of divine dependent on the quantum of proteins, carbohydrates and fats in
the cocoa product and its corresponding digestibility coefficient
* Corresponding author. Tel.: +234 8035782925. (Minife, 1989). Cocoa powder is mainly used for low calorie food
E-mail address: [email protected] (E.I. Adeyeye). products. The minerals present in cocoa powder are important

0308-8146/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.foodchem.2009.04.127
358 E.I. Adeyeye et al. / Food Chemistry 118 (2010) 357–363

for their nutritional value. Olaofe, Oladeji, and Ayodeji (1987) and (PCCS) was a product that resulted from further processing of the
Olaofe and Onajeta (1986) reported on the quality parameters of P-LCN. The processes involve microwave heating of the beans at
cocoa beans from Nigeria, as well as on cocoa-based beverages of a temperature range of 90–100 °C for a period of about 15 min
different brands consumed in the Nigerian market. The level of on a vibratory bed (this makes the cocoa bean shell puff for easy
fermentation, degree of alkalisation, roasting and fat content deter- winnowing), automated roasting (at temperature range of 90–
mine the colour and flavour of cocoa products. Fermentation helps 100 °C for about 20 min in a rotary evaporator), refining to over
to generate proper aroma and reduce the level of acetic acid, which 98% fineness to obtain cocoa liquor (masse) which is further
causes off-flavour in chocolate. The pH of cocoa liquors prepared heat-treated at 80–90 °C for about 12 h in storage tanks. The liquor
from well fermented and dried West African beans is around 5.5 is later fed in batches of about 200 kg sizes into a steam-jacketed
whilst those of unfermented or poorly fermented beans are 5.0 vacuum mixer, where liquor homogenisation and further heating
or less (The Biscuit, Cake, Chocolate, and Confectionary Alliance takes place for about 10 min before final pressing to obtain cocoa
(BCCCCA), 1996). Cocoa is added to cigarettes for flavour enhance- butter and cocoa cake. The final heating and homogenisation are
ment. It also contains various psychoactive compounds, such as used to take the liquor from about 80 °C to about 105 °C and to en-
theobromine, caffeine, serotonin, histamine, tryptophan, tryta- sure maturation of the liquor. This also further ensures a final rapid
mine, tyramine, phenylethylamine, octopamine and anandamide evaporation of residual moisture to <1% in the cocoa liquor and
(Rambali et al., 2002). The levels of these compounds in added guarantees acceptable sterilisation of the liquor. The pressed cake
cocoa in cigarettes are thus critical to curtail possible addiction is kibbled mechanically to obtain smaller sizes of the cocoa cake
to cigarette smoking. Theobromine and theophylline, as well as solid, otherwise called processed cocoa cake samples (PCCS). The
caffeine, all found in this plant, are used as a diuretic, stimulant factory samples were labelled process-line cocoa nibs (P-LCN)
and also, in modern medicine, as an antiasthmatic (Morgan, 1994). and processed cocoa cake samples (PCCS), respectively. These
There is, at present, scanty information on the amino acid pro- formed the group two samples.
file of cocoa beans and the effect of farm and industrial processing
on their relative concentrations. This study attempts to evaluate 2.2. Sample treatment
the amino acid composition of dried unfermented and fermented
cocoa beans from a major cocoa-producing town in south-west The samples were homogenised and ground to fine powder,
Nigeria vis-à-vis that of the in-process cocoa nibs and processed using a Mouliness blender. The ground portions were kept in plas-
cocoa cake from a major cocoa processing industry in Nigeria. tic rubbers in the freezer ( 4 °C) pending analysis. The values re-
The effects of fermentation and heat treatment during the course ported for each test were averages of two or more determinations.
of milling are to be evaluated. Direct processing of the same batch
of beans collected from the selected farm location could not be 2.3. Determination of crude protein
conducted because of the enormous quantity required for batch
processing. Consequently, heterogeneous samples of beans sup- Nitrogen was determined by the micro-Kjeldahl method, as de-
plied, from different locations (thoroughly mixed together to a rep- scribed by Pearson (1976) and the percentage nitrogen was con-
resentative sample), were used for evaluating the effect of heat verted to crude protein by multiplying by 6.25.
treatment during milling. This means the analyses were on four
samples in two major groups; they are fermented and unfer- 2.4. Determination of amino acids
mented samples from the same farm in Aisegba Ekiti and this is
the Forastero Amazonian Group (Opeke, 1992); the second group The amino acids profile in the cocoa samples was determined
is the factory sample group consisting of the process-line cocoa using methods described by Adeyeye and Afolabi (2004). The cocoa
nibs (P-LCN) and the processed cocoa cake samples (PCCS). The fac- samples were dried to constant weight. The mass was subse-
tory samples (P-LCN and PCCS) came from a blend of cocoa beans quently defatted, hydrolysed, filtered to remove the humins and
from different sources of the same species, which is the Forastero evaporated to dryness at 40 °C under vacuum in a rotary evapora-
Amazonian Group. tor. Each residue was dissolved with 5 ml of acetate buffer (pH 2.0)
and stored in a plastic specimen bottle and kept inside the deep
freezer pending subsequent analysis. The Technicon Sequential
2. Materials and methods Multisample Amino Acid Analyser (TSM), Technicon Instruments
Corporation, New York was used for the analysis. The principle is
2.1. Materials based on ion-exchange chromatography (IEC) (FAO/WHO, 1991).
The equipment is designed to separate free acidic, neutral and ba-
Cocoa bean seeds were collected from fully ripe pods harvested sic acids of the hydrolysate. The amount loaded for each sample
from some cocoa trees in a farm plantation located at Aisegba Ekiti was 5–10 ll and about 76 min elapsed for each analysis. The col-
in Ekiti State of Nigeria in December 2007, during the main crop umn flow rate was 0.50 ml/min at 60 °C with reproducibility con-
season. The harvested beans were divided into two portions imme- sistent within ±3%. The net height of each peak produced by the
diately after they were taken out of the pods. Whilst the first por- chart record of the TSM was measured and calculated for the ami-
tion was directly sun-dried in open air, the second portion was no acid it was representing. The averages of two determinations
fermented by a heaping method, using plantain leaves to cover were reported. All chemicals used were of analytical grade.
the beans for 6 days before sun-drying. The dried beans were de-
shelled, dry-milled and labelled as unfermented and fermented 2.5. Estimation of quality of dietary protein
nibs, respectively. These formed the group one samples. Similarly,
in-process cocoa beans (heterogeneous) of indefinite source, and The essential amino acid score was calculated using the follow-
kibbled cake were collected from the production floor of Co-oper- ing formula (FAO/WHO, 1973): amino acid score = amount of ami-
ative Cocoa Products Limited, Akure, Ondo State, Nigeria in Decem- no acid per test protein (mg/g)/amount of amino acid per protein in
ber 2007, for comparative analysis. reference pattern (mg/g).
The process-line cocoa nibs (P-LCN) were prepared from Amino acid score (for both essential and non-essential amino
blended, cleaned and destoned dried cocoa beans from the fac- acids) was also calculated based on whole hen’s egg (Paul, South-
tory’s cleaner/destoner machine. The processed cocoa cake sample gate, & Russel, 1976).
 E.I. Adeyeye et al. / Food Chemistry 118 (2010) 357–363 359

The ratio of total essential amino acid (TEAA) to the total amino Table 2
acid (TAA), i.e. (TEAA/TAA), total sulphur amino acid (TSAA), per- Amino acid profiles (mg/g crude protein) of process-line cocoa nibs and processed
cocoa cake samples from a processing industry.
centage cystine in TSAA (%Cys/TSAA), total aromatic amino acid
(TArAA), and the Leu/Ile ratios were calculated whilst the predicted Amino acid P-LCNa PCCSb Mean SD CV%
protein efficiency ratio (P-PER) was determined using one of the Lys 181 ± 0.05 52.7 ± 0.03 117 90.7 77.5
equations developed by Alsmeyer, Cunningham, and Happich His 11.7 ± 0.01 4.4 ± 0.01 8.05 5.16 64.1
(1974), i.e. P-PER = 0.468 + 0.454 (Leu) 0.105 (Tyr). Arg 46.3 ± 0.02 29.4 ± 0.20 37.9 12.0 31.7
Asp 37.4 ± 0.03 43.7 ± 0.20 40.6 4.45 11.0
Calculations completed were the grand mean, standard devia- Thr 15.9 ± 0.02 8.7 ± 0.01 12.3 5.09 41.4
tion, coefficients of variation in percentage, correlation coefficient, Ser 21.4 ± 0.00 23.1 ± 0.02 22.3 1.20 5.38
regression and F-test, setting the confidence level at 95% (Christian, Glu 37.4 ± 0.10 43.5 ± 0.02 40.5 4.31 10.6
1980). Pro 28.1 ± 0.02 25.1 ± 0.01 26.6 2.12 7.97
Gly 2.9 ± 0.01 5.0 ± 0.01 3.95 1.48 37.5
Ala 1.5 ± 0.01 20.1 ± 0.02 10.8 13.2 122
3. Results and discussion Cys 15.3 ± 0.02 10.8 ± 0.01 13.1 3.18 24.3
Val 39.2 ± 0.03 2.4 ± 0.01 20.8 26.0 125
The crude protein levels of the unfermented and fermented Met 4.3 ± 0.01 3.7 ± 0.02 4.0 0.42 10.5
Ile 63.3 ± 0.30 31.9 ± 0.05 47.6 22.2 46.6
samples are shown in Table 1. Also, the crude protein levels in
Leu 47.2 ± 0.02 37.0 ± 0.20 42.1 7.21 17.1
the process-line cocoa nibs (P-LCN) and processed cocoa cake sam- Tyr 13.6 ± 0.02 12.1 ± 0.11 12.9 1.06 8.22
ples (PCCS) are shown in Table 2. The level of crude protein in fer- Phe 21.8 ± 0.02 14.8 ± 0.01 18.3 4.95 27.0
mented cocoa bean was 15.2 g/100 g which was better than the Try – – – – –
Crude protein (g/100 g) 23.2 ± 0.20 18.4 ± 0.03
value in unfermented cocoa bean (13.6 g/100 g) by 1.58 g/100 g
or 10.4%. This meant that the fermentation process had improved a
Process-line cocoa nibs.
b
the level of the crude protein in the fermented sample compared Processed cocoa cake samples.
to the unfermented sample. In the case of the group two samples,
P-LCN had a crude protein level of 23.2 g/100 g which was better
than the PCCS sample (18.4 g/100 g) by 4.8 g/100 g or 20.7%. This (cp) (in unfermented cocoa nibs) and 153 mg/gcp (fermented cocoa
meant that the heat involved in the processing reduced the level nibs). Another acidic amino acid, aspartic acid (Asp), occupied the
of protein. This reduction could have been due to the Maillard reac- second position in both samples with values of 100 mg/gcp (unfer-
tions which are an interaction between the carbonyl group of a mented sample) and 82.5 mg/gcp (fermented sample). Cystine
reducing sugar and the free amino acid group from an amino acid (Cys) was the least concentrated, in both samples, with values of
or protein. The resulting condensation product is converted by the 7.8 mg/gcp (unfermented cocoa nibs) and 6.9 mg/gcp (fermented
Amadori rearrangement to the 1-deoxy-2-ketosyl compound. nibs). The fermented cocoa nibs were richer than were unfer-
Browning then proceeds along complex pathways, the exact se- mented cocoa nibs in the following amino acids: Lys, His, Arg, Ser,
quence being dependent on pH, temperature, concentration and Glu, Gly, Ala, Val, Ile, Tyr and Phe, whereas Pro (or one amino acid,
the identity of the reactants (Muller & Tobin, 1980). However, 5.88%) was of equivalent level (12.5 mg/gcp in both samples). This
Ala increased in the PCCS. meant that the fermented cocoa nibs were 64.7% richer in the ami-
The amino acids composition of the unfermented and fermented no acids than were the unfermented nibs; whereas the unfer-
cocoa nibs is presented in Table 1. The amino acid concentrations mented sample was only better in five (or 29.4%) of the amino
were variously distributed amongst the two samples, and this could acids. Therefore, fermentation improved the amino acid profile of
easily be seen both in the various samples and in the coefficient of the cocoa nibs. This is particularly true for most of the essential
variation percentage (CV%). Glutamic acid (Glu) was the most con- amino acids: Lys, His, Arg, Thr, Val, Ile and Phe. The improvement
centrated amino acid amongst the samples: 128 mg/g crude protein of amino acid concentration by fermentation followed the trend ob-
served in guinea corn, where steeping of the grains improved the
amino acid profile over the raw and germinated samples in Arg,
Table 1 His, Met, Phe, Thr, Val, Ala, Cys, Gly, Pro, Ser, and Tyr (Adeyeye,
Amino acid profiles (mg/g crude protein) of unfermented and fermented cocoa nibs
from a farm location.
2008).
Our results, in both the unfermented and fermented cocoa nibs,
Amino acid Unfermented Fermented Mean SD CV% followed the trends in Cola acuminata, Garcinia kola and Anacardi-
nibs nibs
um occidentale, where Glu was the most concentrated amino acid
Lys* 42.0 ± 0.02 52.6 ± 0.02 47.3 7.50 15.9 and Asp was the second most concentrated in C. acuminata and
His* 20.0 ± 0.00 23.3 ± 0.02 21.7 2.33 10.7
G. kola (Adeyeye, Asaolu, & Aluko, 2007). Our trend in Glu and
Arg* 43.6 ± 0.01 51.4 ± 0.20 47.5 5.52 11.6
Asp 100 ± 0.10 82.5 ± 0.11 91.3 12.4 13.6 Asp agreed with the results of Olaofe, Adeyemi, and Adediran
Thr* 29.9 ± 0.03 23.3 ± 0.10 26.6 4.67 17.6 (1994) who found that Glu and Asp, respectively, were the first
Ser 23.7 ± 0.01 32.6 ± 0.03 28.2 6.29 22.3 and second most concentrated amino acids in some oilseeds. The
Glu 128 ± 0.20 153 ± 0.40 141 17.7 12.6
differences in the levels of the amino acids in the two samples
Pro 12.5 ± 0.02 12.5 ± 0.03 12.5 0.00 –
Gly 20.5 ± 0.01 32.0 ± 0.02 26.3 8.13 30.9
were shown by the various levels of the CV%; when subjected to
Ala 29.8 ± 0.20 40.1 ± 0.03 35.0 7.28 20.8 the F-test the differences were not significant at p < 0.05 since Fc
Cys 7.8 ± 0.01 6.9 ± 0.02 7.35 0.64 8.71 (1.11) < Ft (2.35). Also, the correlation coefficient was high at r0.96
Val* 32.1 ± 0.10 35.1 ± 0.02 33.6 2.12 6.31 and regression (Rc) was 3.51.
Met* 9.9 ± 0.01 8.0 ± 0.00 8.95 1.34 15.0
Table 2 contains the amino acid profiles for the process-line co-
Ile* 21.4 ± 0.02 29.3 ± 0.20 25.4 5.59 22.0
Leu* 72.2 ± 0.30 62.4 ± 0.20 67.3 6.93 10.3 coa nibs (P-LCN) and processed cocoa cake samples (PCCS). In both
Tyr 18.6 ± 0.02 27.0 ± 0.01 22.8 5.94 26.1 samples, Lys was the most concentrated amino acid with values of
Phe* 28.6 ± 0.01 36.3 ± 0.02 32.5 5.44 16.7 181 mg/gcp in P-LCN and 52.7 mg/gcp in PCCS; whilst Ile (63.3 mg/
Try* –a – – – – gcp) was the second most concentrated in P-LCN, it was Asp
Crude protein (g/100 g) 13.6 ± 0.30 15.2 ± 0.21
(43.7 mg/gcp) in PCCS. The following amino acids were more con-
*
Essential amino acids. centrated in P-LCN than in PCCS (mg/gcp): Lys, His, Arg, Thr, Pro,
a
Not determined. Cys, Val, Met, Ile, Leu, Tyr and Phe, or 70.6% better in amino acid
360 E.I. Adeyeye et al. / Food Chemistry 118 (2010) 357–363

concentration. This meant that P-LCN would be a better protein D-amino acids which are both due to high and prolonged heat

food ingredient than would PCCS. Here some of CV% levels were treatment (Fennema, 1985; Muller & Tobin, 1980).
well above 50.0 and the data, when subjected to F-test analysis, Various parameters are presented in Tables 3 and 4. The total
showed that significant differences existed at p < 0.05 between amino acids (TAA) in unfermented cocoa nibs was 641 mg/gcp
the P-LCN and PCCS amino acid profiles, since Fc (6.86) > Ft and it was 708 mg/gcp in fermented cocoa nibs (Table 4) whilst
(2.35). The r0.70 was less than the value obtained in Table 1 but it was 635 mg/gcp in P-LCN and 368 mg/gcp in PCCS (Table 3).
the Rc was 12.4 which was much higher than the value in Table 1. The TAA in unfermented cocoa nibs was close to the value of
It is interesting to note the high differences in the amino acids, 659 mg/gcp in A. occidentale and also the value of TAA in PCCS
Lys, His, Arg, and Cys between P-LCN and PCCS samples. The avail- was close to the value of 356 mg/gcp in C. acuminaa (Adeyeye
ability of some amino acids, for example, Lys, Met, Arg, Try, Cys and et al., 2007). The level of TAA in fermented cocoa nibs was close
His, is often severely impaired when the protein in the food is to the values of 703–917 mg/gcp of dehulled samples of African
heated, e.g. in processing, or where it is improperly stored. This yam bean (Adeyeye, 1997). The total non-essential amino acids
impairment occurs when the Amadori rearrangement goes beyond (TNEAA) for the samples were (mg/gcp): 341 (unfermented nibs),
the deoxy-ketosyl stage due to heat treatment. This is particularly 421 (fermented nibs) – see Table 3, 158 (P-LCN) and 183 (PCCS)
serious when intravenous drip fluids containing sugars and pro- – see Table 4. The TNEAA of 341–421 mg/gcp was close to the value
teins undergo Maillard reactions during sterilisation (Muller & of 327–454 mg/gcp in African yam bean (Adeyeye, 1997) and also
Tobin, 1980). Carpenter has used the susceptibility of Lys to heat to 323 mg/gcp in A. occidentale (Adeyeye et al., 2007). However, in
damage in the presence of moisture as a basis for estimating the the composition of the total essential amino acids (TEAA), there
extent of the damage (Muller & Tobin, 1980). Although Val may was a reversal of concentration which followed this pattern (mg/
be thermally stable, it may also be possible that it takes part in this gcp, with His): 478 (P-LCN) > 300 (unfermented nibs) > 287 (fer-
type of browning reaction which will definitely reduce its concen- mented nibs) > 185(PCCS). On a percentage basis, this trend was
tration during heat processing. not consistent amongst the samples. The percent TNEAA ranged
The processed cocoa cake samples (PCCS) underwent various between 53.3% (unfermented nibs) and 59.5% (fermented nibs)
heat processing unit operations which lasted for 13 h or more. This with a low value of CV% (7.77) (Table 3); from Table 4 it ranged
would have led to a very serious heat effect on the amino acids. from 24.8% (P-LCN) to 49.8% (PCCS) with a high value of CV%
Looking critically at Table 2, where the values of P-LCN and PCCS (64.8). The % TEAA (with His) ranged between 75.2% (P-LCN) and
are compared, wide variation existed between the essential amino 50.2% (PCCS) with CV% of 28.2 (Table 4). These results showed that
acids of the two samples. For example, loss of amino acid concen- the industrial processed cake was again lower in the TEAA than in
tration from P-LCN to PCCS goes thus (mg/g crude protein) Lys, 128 the process on line by a wide margin. The total neutral amino acids
(78.7%); His, 7.3 (62.4%); Arg, 16.9 (36.6%); Thr, 7.2 (45.3%); Cys, (TNAA) levels were close in (mg/gcp); 228 (unfermented nibs), 235
4.5 (29.4%); Val, 36.8 (90.9%); Met, 0.6 (1.40%); Ile, 31.4 (49.6%); (fermented nibs) but low in P-LCN (74.8) and PCCS (87.2).
Tyr, 1.5 (11.0%) and Phe, 7.0 (32.1%). These values show serious The TEAA in melon and gourd oilseeds with respective values of
reductions in the available essential amino acids, unlike the non- 534 mg/g and 536 mg/gcp (Olaofe et al., 1994) were reportedly
essential amino acids of the PCCS. Usually, the method used for higher than all of our values, that ranged between 185 mg/gcp
the amino acid analysis will only detect L-amino acids from animal and 478 mg/gcp; with the exception of 478 mg/gcp (P-LCN), all of
and plant proteins that do not produce racemisation (White et al., our EAA values were lower than those in soybean (444 mg/gcp)
1973). The reasons for these serious reductions of the amino acids (Kuri, Sundar, Kahuwi, Jones, & Rivett, 1991). Our present samples
from P-LCN to PCCS would likely be due to Amadori rearrangement were either close to or lower than the following TEAA levels (mg/
that goes beyond the deoxy-ketosyl stage and the formation of gcp): pigeon pea (452) (Nwokolo, 1987), pumpkin seed (396)

Table 3
Concentrations of essential, non-essential, acidic, neutral, sulphur, aromatic (mg/g crude protein) of unfermented and fermented cocoa nibs.

Amino acid Unfermented Nibs Fermented nibs Mean SD CV%

Total amino acid (TAA) 641 708 675 43.4 6.43
Total non-essential amino acid (TNEAA) 341 421 381 56.6 14.9
Total essential amino acid (TEAA)
–With His 300 287 294 9.19 3.13
–No His 280 263 272 12.0 4.41
% TNEAA 53.3 59.5 56.4 4.38 7.77
% TEAA
–With His 46.8 40.5 43.7 4.5 10.2
–No His 43.6 37.2 40.4 4.53 11.2
Total neutral amino acid (TNAA) 307 346 327 27.6 8.44
% TNAA 47.9 48.8 48.4 0.64 1.32
Total acidic amino acid (TAAA) 228 235 232 4.95 2.13
% TAAA 35.6 33.2 34.4 1.7 4.94
Total basic amino acid (TBAA) 106 127 117 14.8 12.6
% TBAA 16.5 18.0 17.3 1.06 6.13
Total sulphur amino acid (TSAA) 17.7 14.9 16.3 1.98 12.1
% TSAA 2.76 2.11 2.44 0.46 18.9
% Cys in TSAA 44.1 46.3 45.2 1.56 3.45
Total aromatic amino acid (TArAA) 47.2 63.3 55.3 11.4 20.6
% TArAA 7.36 8.94 8.15 1.12 13.7
P-PER* 3.55 2.55 3.05 0.71 23.3
Leu/Ile ratio 3.37 2.13 2.75 0.88 32.0
Leu-Ile (difference) 50.8 33.1 42.0 12.5 29.8
% Leu-Ile 70.4 53.0 61.7 12.3 19.9
*
Predicted protein efficiency ratio.
 E.I. Adeyeye et al. / Food Chemistry 118 (2010) 357–363 361

Table 4
Concentrations of essential, non-essential, acidic, neutral, sulphur, aromatic amino acids (mg/g crude protein) of P-LCN and PCCS.

Amino acid P-LCN* PCCS* Mean SD CV%

Total amino acid (TAA) 635 368 502 189 37.6
Total non-essential amino acid (TNEAA) 158 183 171 17.7 10.4
Total essential amino acid (TEAA)
–With His 478 185 332 207 62.3
–No His 466 181 324 202 62.3
% TNEAA 24.8 49.8 27.3 17.7 64.8
% TEAA
–with His 75.2 50.2 62.7 17.7 28.2
–no His 73.3 49.0 61.2 17.2 28.1
Total neutral amino acid (TNAA) 322 195 259 89.8 34.7
% TNAA 50.7 52.9 51.8 1.56 3.01
Total acidic amino acid (TAAA) 74.8 87.2 81.0 8.77 10.8
% TAAA 11.8 23.7 17.8 8.4 47.2
Total basic amino acid (TBAA) 239 86.5 163 108 66.3
% TBAA 37.6 23.5 30.6 9.97 32.6
Total sulphur amino acid (TSAA) 19.6 14.5 17.1 3.61 21.1
% TSAA 3.09 3.94 3.52 0.60 17.0
% Cys in TSAA 78.1 74.5 76.3 2.55 3.34
Total aromatic amino acid (TArAA) 35.4 26.9 31.2 6.01 19.3
% TArAA 5.57 7.31 6.44 1.23 19.1
P-PER* 2.47 2.02 2.25 0.32 14.2
Leu/Ile ratio 0.75 1.16 0.96 0.29 30.2
Leu-Ile (difference) 16.1 5.1 10.6 7.78 73.4
% Leu-Ile 34.1 13.8 24.0 14.4 60.0
*
See Table 2.

(Aisegbu, 1987), cowpea (426) (Olaofe, Umar, & Adediran, 1993) The predicted protein efficiency ratios (P-PER) were better in
and Cajanus cajan (436) (Oshodi, Olaofe, & Hall, 1992). This meant the unfermented (3.55) and fermented (2.55) cocoa nibs than in
that unfermented nibs, fermented nibs and PCCS protein in the the P-LCN (2.47) and PCCS (2.02) samples. The experimentally
samples were of lower quality than those in cowpea, soybean determined PER usually ranged from 0.0 for a very poor protein
and pigeon pea. However, whilst Cys was 0.0 mg/gcp in melon, to a maximum possible of just over 4 (Muller & Tobin, 1980). These
pumpkin seed and gourd seed (Olaofe et al., 1994) and 11.3 mg/ results showed that P-LCN and PCCS would likely be less utilised in
gcp in A. occidentale, 2.5 mg/gcp in G. kola and 4.5 mg/gcp in C. the body than would the other two samples.
acuminata (Adeyeye et al., 2007), it was (mg/gcp): in unfermented The Leu/Ile ratio values ranged as follows: 3.37 (unfermented
nibs (17.7), fermented nibs (14.9), P-LCN (19.6) and PCCS (14.5). nibs), 2.13 (fermented nibs), 0.75 (P-LCN) and 1.16 (PCCS). From
Generally, most of our results were better in many of the amino Table 1, the level of Leu was more than twice that of Ile in unfer-
acids (essential and non-essential) than was pumpkin seed (Olaofe mented and fermented nibs whilst, in Table 2, the level of Leu
et al., 1994). was just above one half that of Ile in P-LCN and slightly above Ile
Whilst it is known that cystine can spare part of the require- in PCCS. It has been suggested that an amino acid imbalance from
ment for methionine, FAO/WHO/UNICEF (1985) does not give excess leucine might be a factor in the development of pellagra in
any indication of the proportion of total sulphur amino acids that sorghum consumption (FAO, 1995).
can be met by Cys. For the rat, chick and pig, the proportion is High Leu in the diet impairs tryptophan and niacin metabolism
about 50% (FAO/WHO, 1991). Most animal proteins are low in cys- and is responsible for niacin deficiency in sorghum eaters (Belava-
tine; in contrast, many vegetable proteins, especially the legumes, dy, Srikantia, & Gopalan, 1963) . This leads to the hypothesis that
contain substantially more Cys than methionine. Thus, for animal excess Leu in sorghum is aetiologically related to pellagra in sor-
protein, Cys is unlikely to contribute more than 50% of the total ghum-eating populations (FAO, 1995). The study of Krishnaswamy
sulphur amino acids (FAO/WHO, 1991). For our samples, the per- and Gopalan (1971) suggested that Leu/Ile balance is more impor-
centages of Cys in total sulphur amino acids were: 44.1% (unfer- tant than dietary excess of Leu alone in regulating the metabolism
mented nibs), 46.3% (fermented nibs) – see Table 3; 78.1 (P-LCN) of Try and niacin and hence the disease process. Experiments in
and 74.5 (PCCS) – see Table 4. Whilst the Cys/TSAA% in unfer- dogs have shown that animals fed sorghum proteins (with less
mented and fermented nibs followed the trend in G. kola (37.8%) than 110 mg/gcp of Leu) did not suffer from nicotinic acid defi-
and C. acuminata (44.3%) (Adeyeye et al., 2007), as well as in ciency (Belavady & Udayasekhara Rao, 1979). None of our samples
animals: 36.3% (Macrotermes bellicosus) (Adeyeye, 2005a), 25.6% had levels of Leu up to 110 mg/gcp.
(Zonocerus variegatus) (Adeyeye, 2005b), 35.5% (Archachatina mar- Table 5 contains the amino acid scores, based on the provisional
ginata), 38.8% (Archatina archatina) and 21.0% (Limicolaria sp.) amino acids, for the unfermented and fermented cocoa nibs whilst
(Adeyeye & Afolabi, 2004), the Cys/TSAA% in P-LCN and PCCS fol- Table 6 contains the scores of P-LCN and PCCS, based on the same
lowed the trend in A. occidentale (50.5%) (Adeyeye et al., 2007), formula. In Table 5, the limiting amino acids for both unfermented
coconut endosperm (62.9%) (Adeyeye, 2004), raw guinea corn and fermented cocoa nibs were Met + Cys with respective values of
(58.9%), steeped guinea corn (72.0%) and germinated guinea corn 0.51 (unfermented nibs) and 0.43 for fermented nibs. The entire
(71.1%) (Adeyeye, 2008). This meant that, whilst both unfermented CV% was low and no significant differences existed between the
and fermented cocoa nibs behaved like animal proteins under two samples at p < 0.05 (F-test). Therefore, in order to fulfil the
these conditions, the P-LCN and PCCS behaved like plant proteins. day’s needs for the EAA in unfermented cocoa nibs, 100/51 or
FAO/WHO (1973), states that Cys may supply up to one-third of the 1.96 times as much unfermented nibs would have to be eaten when
need for total sulphur amino acids whilst tyrosine may also supply it is the sole protein in the diet; in fermented nibs, it would be 100/
up to one-third of the need for total aromatic amino acids. 43 or 2.33 times the protein level. In Table 6, Thr was the limiting
362 E.I. Adeyeye et al. / Food Chemistry 118 (2010) 357–363

Table 5 results not significantly different; in TEAA/TNEAA in unfermented

Essential amino acid scores of the unfermented and fermented cocoa nibs based on nibs, Fc > Ft; in fermented nibs, Fc > Ft, in P-LCN, Fc > Ft meaning
provisional amino acid scores.
all the values were significantly different but, in PCCS, Fc < Ft; hence
Amino acid Unfermented nibs Fermented nibs Mean SD CV% results were not significantly different. For the EAA score/EAA
Ile 0.54 0.73 0.64 0.13 20.3 score, in unfermented/fermented nibs, Fc < Ft, but not significant;
Leu 1.07 0.89 0.98 0.13 13.3 for P-LCN/PCCS, Fc > Ft, and results were significantly different at
Lys 0.76 0.96 0.86 0.14 16.3 p < 0.05.
Met + Cys 0.51 0.43 0.47 0.06 12.8
Phe + Tyr 0.79 1.06 0.93 0.19 20.4
In conclusion, the findings of this study showed that there was a
Thr 0.76 0.58 0.67 0.13 19.4 more positive build up of AA in fermented nibs than in unfer-
Try – – – – – mented nibs. In the processing, about 40% of the AA was destroyed;
Val 0.64 0.7 0.67 0.04 5.97 this might have resulted from the series of reactions involving ami-
no acids, nitrates and antioxidants during the heat processing of
cocoa. Finally, more cake (PCCS) would be required for comple-
Table 6 mentation/fortification than would have been used if the unpro-
Essential amino acid scores of the P-LCN and PCCS based on provisional amino acid cessed nibs were to be used. Also the differences observed in the
scores. fermented and unfermented samples (group one) and P-LCN and
Amino acid P-LCN* PCCS* Mean SD CV% PCCS (group two) could be due to planting material, climate, vari-
eties, application of fertiliser, heat treatments and storage between
Ile 1.58 0.80 1.19 0.55 46.2
Leu 0.67 0.53 0.60 0.10 16.7
the group one and two samples.
Lys 3.28 0.96 2.12 1.64 77.4
Met + Cys 0.56 0.41 0.49 0.11 22.4
Acknowledgement
Phe + Tyr 0.59 0.45 0.52 0.11 19.2
Thr 0.40 0.22 0.31 0.13 41.9
Try – – – – – The authors are grateful to Dr. Segun Awolumate, the Managing
Val 0.78 0.05 0.42 0.52 124 Director of Co-operative Cocoa Products Limited, Akure, Nigeria for
his collaborative support.

References
Table 7
Amino acid scores of the four samples based whole hen’s egg. Adeyeye, E. I. (1997). Amino acid composition of six varieties of dehulled African
yam bean (Sphenostylis stenocarpa) flour. International Journal of Food Sciences
Amino acid Unfermented nibs Fermented nibs P-LCN PCCS
and Nutrition, 48, 345–351.
Lys 0.68 0.85 2.92 0.85 Adeyeye, E. I. (2004). The chemical composition of liquid and solid endosperm of
His 0.83 0.97 0.49 0.18 ripe coconut. Oriental Journal of Chemistry, 20(3), 471–476.
Arg 0.71 0.84 0.76 0.48 Adeyeye, E. I. (2005a). The composition of the winged termites, Macrotermes
Asp 0.93 0.77 0.35 0.41 bellicosus. Journal of Chemical Society of Nigeria, 30(2), 145–149.
Thr 0.59 0.46 0.31 0.17 Adeyeye, E. I. (2005b). Amino acid composition of variegated grasshopper,
Zonocerus variegatus. Tropical Science, 45(4), 141–143.
Ser 0.3 0.41 0.27 0.29
Adeyeye, E. I. (2008). The intercorrelation of the amino acid quality between raw,
Glu 1.07 1.28 0.31 0.36
steeped and germinated guinea corn (Sorghum bicolor) grains. Bulletin of the
Pro 0.33 0.33 0.74 0.66 Chemical Society of Ethiopia, 22(1), 11–17.
Gly 0.68 1.07 0.1 0.17 Adeyeye, E. I., & Afolabi, E. O. (2004). Amino acid composition of three types of land
Ala 0.55 0.74 0.028 0.37 snails consumed in Nigeria. Food Chemistry, 85, 535–539.
Cys 0.43 0.38 0.85 0.6 Adeyeye, E. I., Asaolu, S. S., & Aluko, A. O. (2007). Amino acid composition of two
Val 0.43 0.47 0.52 0.032 masticatory nuts (Cola acuminata and Garcinia kola) and a snack nut
Met 0.31 0.25 0.13 0.12 (Anacardium occidentale). International Journal of Food Sciences and Nutrition,
Ile 0.38 0.52 1.13 0.57 58(4), 241–249.
Leu 0.87 0.75 0.57 0.45 Aisegbu, J. E. (1987). Some biochemical evaluation of fluted pumpkin seed. Journal
Tyr 0.47 0.68 0.34 0.3 of the Science of Food and Agriculture, 40, 151–155.
Phe 0.56 0.71 0.43 0.29 Akinyeye, R. O. (1999). Causes of capacity under-utilization in cocoa processing
Try – – – – industries in Nigeria. Akure, Nigeria: MBA Thesis, Federal University of
Technology.
Alsmeyer, R. H., Cunningham, A. E., & Happich, M. L. (1974). Equations to predict
PER from amino acid analysis. Food Technology, 28, 34–38.
amino acid in P-LCN (0.40) and Val in PCCS (0.05); their correction Belavady, B., & Udayasekhara Rao, P. (1979). Leucine and isoleucine content of jowar
values would be 2.50 times the protein in P-LCN and 20 times in and its pellagragenicity. Indian Journal of Experimental Biology, 17, 659–661.
PCCS. There were CV% levels of 124 (Val), 77.4 (Lys) and 46.2 (Ile) Belavady, B., Srikantia, S. G., & Gopalan, C. (1963). The effect of oral administration
of leucine on the metabolism of tryptophan. Biochemical Journal, 87, 652–655.
and significant differences existed between the P-LCN and PCCS Christian, G. D. (1980). Analytical chemistry (3rd ed.). New York, USA: John Wiley
samples at p < 0.05 when subjected to F-test analysis. and Sons Inc.
Table 7 contains the amino acid scores of the four samples FAO (1995). Sorghum and millets in human nutrition.. Rome, Italy: FAO Food
Nutrition Series, No. 27, Food and Agriculture Organisation of the United
based on whole hen’s egg amino acid profile. Serine was the limit-
Nations.
ing amino acid in unfermented nibs (0.30), Met in fermented nibs FAO/WHO (1973). Energy and protein requirements. Geneva, Switzerland: Technical
(0.25), Ala in P-LCN (0.028) and Val in PCCS (0.032). The respective Report Series No. 522, WHO.
corrections would then be 100/30 or 3.3 (unfermented nibs), 100/ FAO/WHO (1991). Protein quality evaluation. Rome, Italy: Report of Joint FAO/WHO
Expert Consultation, FAO Food and Nutrition Paper 51.
25 or 4.0 (fermented nibs), 100/2.8 or 35.7 (P-LCN) and 100/3.2 or FAO/WHO/UNICEF (1985). Energy and protein requirement. Geneva, Switzerland:
31.3 (PCCS) times the protein of each sample where they serve as WHO Technical Report Series No. 724, WHO.
the sole protein sources. Fennema, O. R. (1985). Principles of food science (2nd ed.). New York: Marcel Dekker,
Inc..
The data obtained for TNEAA, TEAA and EAA scores were all International Cocoa and Commodities Organisation (2000). ICCO celebration of cocoa
subjected to the F-test as follows: TNEAA/TNEAA, TEAA/TNEAA 2000. London: K.P. Partners in Publishing.
and EAA score/EAA score for the pairs of unfermented/fermented Kuri, Y. E., Sundar, R. K., Kahuwi, C., Jones, G. P., & Rivett, D. E. (1991). Chemical
composition of Monerdica charantis L. fruits. Journal of Agricultural and Food
nibs and P-LCN/PCCS setting the p < 0.05 (Christian, 1980). The Chemistry, 39, 1702–1703.
following results were obtained. In unfermented/fermented cocoa Minife, B. W. (1989). Chocolate, cocoa and confectionary (3rd ed.). London: Chapman
nibs, TNEAA/TNEAA Fc < Ft and in P-LCN/PCCS result was Fc < Ft, and Hall.
 E.I. Adeyeye et al. / Food Chemistry 118 (2010) 357–363 363

Morgan, J. (1994). Chocolate: A flavour and texture unlike any other. American Opeke, L. K. (1992). Tropical tree crops. Ibadan: Spectrum Books Limited.
Journal of Clinical Nutrition, 60(6), 1065–10675. Oshodi, A. A., Olaofe, O., & Hall, G. M. (1992). Amino acid, fatty acid and mineral
Muller, H. E., & Tobin, G. (1980). Nutrition and food processing. London: Crown Helm. composition of pigeon pea (cajanus cajan). International Journal of Food Sciences
Nwokolo, E. (1987). Nutritional Evaluation of Pigeon Pea meal. Plant Foods for and Nutrition, 42, 187–191.
Human Nutrition, 37, 283–290. Paul, A. A., Southgate, D. A. T., & Russel, J. (1976). First supplement to Mclance and
Olaofe, O., Adeyemi, F. O., & Adediran, G. O. (1994). Amino acid and mineral Widdowson’s the composition of Foods. London, UK: HMSO.
compositions and functional properties of some oil seeds. Journal of Agricultural Pearson, D. (1976). Chemical analysis of foods (7th ed.). London: J.A. Churchhill.
and Food Chemistry, 42, 878–881. Rambali, B., Van Andel, I., Schenk E., Wolterink, G., Van de Werken, G., Stevenson, H.,
Olaofe, O., Oladeji, E. O., & Ayodeji, O. E. (1987). Metal contents of some cocoa beans & Vleenming, W. (2002). The contribution of cocoa additive to cigarette
produced in Ondo State, Nigeria. Journal of the Science of Food and Agriculture, 41, smoking addiction. RIVM report 650270002/2002.
241–244. The Biscuit, Cake, Chocolate, Confectionary Alliance (BCCCCA) (1996). Cocoa beans,
Olaofe, O., & Onajeta, C. C. (1986). Quality assessment of some cocoa-based beverages chocolate manufacturers’ quality requirements (4th ed.). London: BCCCCA.
in the Nigerian market. Nigerian Journal of Nutrition Science, 7(2), 81–85. White, A., Handler, P., & Smith, E. L. (1973). Principles of biochemistry (5th ed.). New
Olaofe, O., Umar, Y. O., & Adediran, G. O. (1993). The effect of nematicides on the York: McGraw-Hill, Inc.
nutritive value and functional properties of cowpea seeds (Vigna unguicalata L.
walp). Food Chemistry, 46(4), 337–342.