Journal of Chromatographic Science, Vol.

47, September 2009

Characterization of Jamaican Agro-Industrial Wastes.

Part I: Characterization of Amino Acids Using HPLC:
Pre-column Derivatization with Phenylisothiocyanate

Y.A. Bailey-Shaw1,*, K.D. Golden1, A.G.M. Pearson1, and R.B.R. Porter2

1Department of Basic Medical Sciences (Biochemistry Section) and 2Department of Chemistry, University of the West Indies, Mona, Kingston
7, Jamaica W.I.

Abstract recent public concerns associated with genetically modified

crops such as soybean and maize (corn), incidents of dioxin con-
Jamaican agro-industries generate large quantities of wastes, tamination of feeds and bovine spongiform encephalopathy
which are either discarded or under-utilized. In order to evaluate (BSE) or mad cow disease, which is said to be linked to contam-
the possible utilization of these wastes, it is necessary that inated meat and bone meal fed to animals, the utilization of
the profiles of the major biochemical groups be developed. cheaper, indigenous feed materials from alternative sources is
This paper describes the determination of the amino acid now becoming highly desirable (2).
composition of coffee, citrus, and rum distillery wastes using Jamaica, being a developing country, is no exception to these
a reversed-phase high-performance liquid chromatography adverse effects. Currently, Jamaica has three animal feed mills
method. Acid hydrolysates of the wastes are derivatized
[Newport Mills, Masterblend, and Jamaica Livestock Association
with phenylisothiocyanate. They are analyzed as their
phenylthiocarbamyl derivatives and determined quantitatively
(JLA) Feed Mills], all utilizing corn and soybean in their manu-
using norleucine as the internal standard. The presence of all facturing processes. However in recent times, the sustained high
the 17 amino acids investigated, nine of which include those importation prices for corn and soybeans, for which there is
essential for animal nutrition, are observed in the samples no domestic production, has resulted in a 12% increase in
investigated, suggesting a high quality of protein with domestic feed prices, which helped to curb the 11.3 and 6.5%
implications in the formulation of animal feeds. expansion in the poultry industry during 2000 and 2001, respec-
tively, to a mere 1.1% during 2002. Due to this escalation in feed
prices, a significant proportion of smaller independent poultry
farmers, who represent ~ 30% of the poultry industry, shifted
Introduction production to a seasonal basis. In addition, there was restricted
growth in mixed feed production, resulting in an expansion in
Agro-industrial wastes contain carbohydrates, proteins, and feed production of less than 1% in 2003 (351,000 MT) and a sim-
lipids, among other biochemicals. However, these wastes are ilar percentage in 2004 (356,000 MT) (3). On this basis, the
often discarded. A strategy of bioremediation is to utilize these search for alternative sources of protein for animal feed is there-
materials as components of animal feeds. Approximately 1000 fore timely.
million tonnes of animal feed is produced globally on an annual The quality of the protein in terms of animal feed is deter-
basis, with feed for poultry produced in the highest tonnage, fol- mined by the amino acid composition and their availability to the
lowed by pig and cattle feeds. Although feed production for aqua- animals. These are usually supplied in the form of protein or
culture is currently relatively low (14 million tonnes), there has crystalline amino acids in the feed (4). However, the require-
been an increasing demand for feed for farmed fish and crus- ments will vary, depending on the species and the age of the ani-
taceans. Protein is the key building block for these feed formula- mals. The objective of this study was, therefore, to characterize
tion systems, with the three main sources being from oilmeals the amino acids in coffee, citrus, and rum distillery wastes in
(316 million tonnes, soybean dominating as a protein source), terms of their composition using high-performance liquid chro-
animal by-products (10 million tonnes), and fishmeal (7 million matography (HPLC).
tonnes) (1). However, due to increasing importation costs, world Determination of amino acid composition has been achieved by
population growth, particularly in developing countries, and the use of several techniques. These include gas chromatography
(5–7), capillary electrophoresis (8), and HPLC (9–30). In recent
* Author to whom correspondence should be addressed: yvonneb@src-jamaica.org times, most amino acid analyses have been conducted using

674 Reproduction (photocopying) of editorial content of this journal is prohibited without publisher’s permission.
Journal of Chromatographic Science, Vol. 47, September 2009

HPLC, in which a universal detection method such as refractive Reagents

index, light scattering, or very low UV wavelengths are used, or All reagents used, unless otherwise stated, were analytical
derivatization with substances that absorb in the visible/UV wave- grade. A kit of individual amino acid standards, HPLC-grade
lengths or fluoresce. Derivatization is done after elution from an methanol and triethylamine (TEA) were obtained from Sigma
HPLC column (post-column derivatization) or prior to HPLC sep- Chemical (St. Louis, MO). HPLC-grade acetonitrile was obtained
aration (pre-column derivatization). Post-column derivatization from Acros (New Jersey, NJ). PITC and ammonium acetate were
methodologies involving the use of reagents such as ninhydrin (9) obtained from Aldrich Chemical (Milwaukee, WI). Water used
and o-phthaladehyde (OPA) (10,11) have been published. The rel- was deionized then purified with a Milli-Q ultrapure water
ative instability of the ninhydrin reagent has however been cited as system (Millipore, Bedford, MA).
a major disadvantage in using this method (12). The remaining
techniques which involve the use of pre-column derivatization HPLC Apparatus
with phenylisothiocyanate (PITC) (13–16), 6-aminoquinolyl-N- HPLC analyses were performed using a Beckman System
hydroxysuccinimidyl carbamate (AQC) (17,18), OPA (19–22), Gold-Nouveau HPLC Unit (Beckman Instruments, Fullerton,
dimethylaminoazobenzenesulfonyl chloride (DABSYL Chloride) CA) equipped with a 126 Programmable Solvent Module with
(23,24 ), dimethylaminonaphthalenesulfonyl chloride (DANSYL a binary pump, a 168 Photodiode Array Detector, a 508
Chloride) (25, 26), 9-fluorenylmethyl chloroformate (FMOC-Cl) Autosampler fitted with a 20 µL loop and utilizing Gold Nouveau
(27, 28) and 4-fluoro-7-nitrobenzo-2-oxa-1,3-diazole or NBD-F Software. Chromatographic separation was performed on an
(29, 30) have also been published. analytical reverse phased column (Waters Spherisorb ODS2, 5
Pre-column derivatization procedures are generally faster µm, 250 × 4.6 mm) at room temperature (26°C).
and more sensitive than post-column derivatization (31). None-
theless, these applications have been limited by several factors. Sample Preparation
These include: the lack of reactivity with secondary amines such Prior to extraction, solid samples (coffee pulp and citrus juice
as proline, instability of the fluorescent products, and difficulties extractor residues) were freeze dried and stored in sealed plastic
in quantitation due to their sensitivity to quenchers in the bags in a dessicator at 0–4°C.
case of OPA (13,14). The lack of selectivity (reacts with both Citrus press liquor, citrus wash water, coffee pulping water,
hydroxyl and amine groups), long reaction times, and high reac- and wash water containing mucilage were stored at –10°C
tion temperatures required by dansyl chloride and the need to without prior preparation. The distillery wastes, on the other
use excess FMOC-Cl reagent that must be extracted prior to hand, were separated into fractions of dissolved (supernatant)
chromatography (often results in hydrolysis and loss of and suspended solids (pellets) by centrifugation procedures prior
FMOC–amino acid adducts) (13), are but a few of the limitations to analysis. Both fractions were stored at 4°C.
of pre-column derivatization. However, the choice of any one Reference to “solid wastes” from here on will relate to coffee
technique is ultimately determined by the sensitivity required pulp, citrus juice extractor residues, and suspended solids of dis-
from the assay. PITC was on this basis the reagent of choice for tillery wastes, whereas “liquid wastes” will be used to refer to
use in this study because it reacts with both primary and sec- press liquor, washing water (coffee and citrus), and dissolved
ondary amino acids to yield stable phenylthiocarbamyl (PTC) solids of distillery wastes.
derivatives that can be detected by UV absorption at 254 nm. In
addition, the analysis is rapid, sensitive, and simple to quantify, Protein extraction from solids
overcoming the disadvantages of detection using dansyl chlo- Total soluble protein was extracted from samples of solids
ride, OPA, among others. using 20 mM sodium phosphate buffer (pH 7) containing 0.1%
Triton-X-100 (TX-100). Approximately 0.5 g of freeze dried
sample was homogenized for 30 s at high speed with 15 mL of
chilled extractant in a Waring laboratory blender pre-chilled to ~
Experimental 4°C. The homogenate was then sonicated (Heat Systems
Ultrasonics Inc. Sonicator/Cell Disruptor, Model W 220F fitted
Samples with a microtip) for 10 min at 60 W using pulses of 45 s while
Coffee processing wastes [coffee pulp, coffee pulping water keeping the tube immersed in an ice bath. The sonicated sample
(water that takes the pulp from the pulper) and wash water con- was then centrifuged at 27,200 × g, for 30 min at 4°C (Beckman
taining mucilage (water that takes the mucilage from the aqua- J2-21 Centrifuge, JA – 20 rotor). The supernatant was decanted
pulper)] were obtained from a coffee factory (traditional and the homogenization, sonication, and centrifugation proce-
aquapulper, wet processing) in St. Andrew, Jamaica. dures repeated using the pellet. The pellet was washed twice only
Citrus wastes [juice extractor residues-orange and grapefruit where possible. The supernatants obtained were pooled, and the
pulp, wash water (water used to wash fruits and floors) and press combined solution used as the soluble protein extract.
liquor (waste from pressing of juice extractor residues during
production of citrus meal)] were obtained from a commercial Protein extraction from liquids
citrus processing plant in St. Catherine, Jamaica. Total soluble proteins were extracted from liquid wastes using
Liquid distillery wastes [pot stills and fermentor bottoms 0.15 M sodium chloride. Approximately 5 mL of liquid sample
(light and heavy) and continuous still effluent] were obtained was diluted with 5 mL of 0.15 M solution of sodium chloride. The
from a sugar factory/rum distillery in Trelawny, Jamaica. resulting mixture was then sonicated for 1 min at 60 W using

675
 Journal of Chromatographic Science, Vol. 47, September 2009

pulses of 45 s. The sonicated sample was then incubated with 1 stream of N2 was used to remove the solvents and the tube was
mL of 0.1% TX-100 at room temperature (26°C) for 30 min. This closed and stored at 4°C prior to analysis. Prior to injection, 300
final solution was used as the soluble protein extract. µL of 0.05 M ammonium acetate was added to each tube.

Hydrolysis and derivatization of amino acids Recovery of amino acids

Soluble protein extracts were hydrolyzed according to the The efficiencies of the extraction, hydrolysis, and derivatiza-
method of González-Castro et al. (13) with minor modifications. tion procedures were investigated by replacing 1 mL of the
Approximately 6 mL of 12 N HCl was added to 6 mL of sample extractant with 1 mL of 5.88 µmol/mL standard amino acid mix-
extract in a 16 × 125 mm, teflon-lined screw cap Pyrex tube, to ture (spiking), extracting total soluble proteins, and then sub-
produce a solution with a final concentration of 6 N. The tube jecting the extracts to hydrolysis and derivatization as described
was thoroughly flushed with nitrogen (N2), quickly capped, and earlier.
placed in an oven at 110°C for 24 h. The contents of the tube Recoveries were calculated based on the difference between
were vacuum filtered, (Whatman No. 1 filter paper, Whatman the total amount determined in the spiked and that in the sam-
International Ltd., Kent, U.K.) to remove solids after hydrolysis. ples extracted without the addition of amino acids (non-spiked).
Aliquots of the solution were further filtered through 0.45-µm
pore-size membrane (Millipore Corp., Bedford, MA). HPLC analysis
Derivatization was carried out according to the method of A step-wise gradient elution program developed in our labora-
González-Castro et al. (13) to which minor modifications were tory was used. Solvent A consisted of an aqueous solution of 0.05
made. A standard mixture containing 2.5 µmol/mL of each M ammonium acetate and Solvent B was 0.1 M ammonium
amino acid in 0.10 N HCl was prepared. Twenty microliters (20 acetate in acetonitrile–methanol–water (44:10:46), both
µL) of this standard or 100 µL of hydrolyzed sample extract were adjusted to pH 6.8 with phosphoric acid (10). The profile was as
then mixed with 10 µL of norleucine (10 µmol/mL) in an follows: 0–0.2 min, 0% B; 0.2–35 min, 50% B; 35–40 min, 75%
Eppendorf tube. The norleucine was used as an internal B; 40–45 min, 75% B; 45–55 min, 0% B, followed by the next
standard. The mixture was then dried in vacuo using a injection. The flow rate was 1 mL/min, and the detector was set
Savant Speedvac. Sixty microliters (60 µL) of methanol– at a wavelength of 254 nm. Twenty microliters (20 µL) of each
water–triethylamine (TEA) (2:2:1, v/v) was added to the residue, sample was injected for analysis. All HPLC determinations were
the contents of the tube were then mixed and dried in vacuo. conducted in duplicate.
Sixty microliters (60 µL) of the derivatizing reagent,
methanol–water–TEA–PITC (7:1:1:1, v/v) was added, the tube
was agitated and left to stand at room temperature for 20 min. A
Results and Discussion

The elution profile of the PTC standard amino acid mixture

showed a good separation of each component (Figure 1).
PITC–TEA-derived components were detected in both standard
and samples; however, these components did not cause any

Figure 1. Chromatogram of PTC standard amino acids. Peak labels are as fol-
lows: asp, aspartic acid; glu, glutamic acid; oh-pro, hydroxy-proline; ser,
serine; gly, glycine; thr, threonine; ala, alanine; his, histidine; pro, proline;
arg, arginine; tyr, tyrosine; val, valine; met, methionine; isl, isoleucine; leu,
leucine; nl, norleucine; phe, phenylalanine; lys, lysine; bp, by-products Figure 2. Chromatogram of PTC amino acids in continuous still effluent sus-
(PITC-TEA derivatives). pended solid (pellet). Peak labels are the same as Figure 1.

676
Journal of Chromatographic Science, Vol. 47, September 2009

interference in the detection of the PTC amino acids. These Representative chromatograms of some liquid and solid waste
peaks were observed to elute near arginine and after all the other samples are illustrated in Figures 2 and 3.
PTC-amino acids had eluted. González-Castro et al. (13) and The content of each amino acid was calculated using the nor-
Heinrikson and Meredith (14) made similar observations. leucine internal standard addition method, and the percentage
However, the latter authors suggested that distillation of the recoveries calculated by spiking duplicate samples with a stan-
pyridine and triethylamine would reduce the number and extent dard mixture of amino acids and then subjecting them to extrac-
of these peaks. Similar profiles were obtained for sample extracts. tion, hydrolysis, and derivatization procedures.
Tables I–IV list the amino acid contents and percentage recov-
eries for all wastes examined. Of importance is the fact that all
amino acids essential for animal nutrition: phenylalanine, histi-
dine, arginine, leucine, threonine, lysine, methionine,
isoleucine, and valine (32), with the exception of tryptophan,
which was completely destroyed during acid hydrolysis (33),
were detected in the samples analyzed, suggesting a high quality
of protein. However, it should be noted that due to variations in
digestion and absorption, the ration which has the highest pro-
tein quality is usually the one that supplies all the essential
amino acids needed, in proportions similar to those in which
they exist in the protein to be formed (34).
Continuous still suspended solids effluent had the highest
total amino acid content of all the solid wastes (119.8 mg/100 g),
while heavy fermentor bottom suspended solids effluent was
found to have the lowest content of total amino acid (16.4
mg/100 g). With the exception of coffee pulp (43.8% of the total
amino acid detected were essential amino acids), more than 50%
of total amino acids detected in the solid waste samples were
Figure 3. Chromatogram of PTC amino acids in continuous still effluent
dissolved solids (supernatant). Peak Labes are the same as Figure 1. amino acids essential to animal nutrition. Orange pulp, which
had a total amino acid content of 83.7 mg/100 g, contained the
highest percentage (71.3%) of essential amino
Table I. Amino Acid Content and Recoveries of Coffee Processing Wastes as acids.
Determined by RP- HPLC of PTC-Amino Acids* On the other hand, analysis of data with
respect to liquid wastes revealed that heavy pot
Coffee Pulp Wash Water still effluent (dissolved solids) contained the
Amino mg/100 g Recovery Pulping Recovery with Mucilage Recovery highest total amino acid content (49.7 mg/L).
Acid Dry Weight (%) Water (mg/L) (%) (mg/L) (%) Citrus press liquor was found to contain the
Asp 5.1 ± 0.2 97.0 0.81 ± 0.06 99.1 1.45 ± 0.02 97.9 lowest content of total amino acid (3.20 mg/L).
Glu 3.9 ± 0.2 96.6 2.0 ± 0.1 99.1 3.20 ± 0.02 97.5 All liquid wastes, with the exception of pot stills:
Phe† 1.5 ± 0.4 96.8 0.55 ± 0.02 99.1 0.180 ± 0.001 97.2 heavy and light, and heavy fermentor bottom
Gly 0.99 ± 0.05 97.0 1.06 ± 0.09 99.3 1.16 ± 0.08 97.7 effluents (39.7%, 45.3%, and 45.0% essential
His† 5.9 ± 0.1 94.4 4.1 ± 0.2 98.6 7.11 ± 0.04 96.3 amino acids, respectively), were found to contain
Arg† 0.66 ± 0.09 98.0 1.89 ± 0.09 99.5 2.1 ± 0.2 98.4 greater than 50% of their total amino acids
Pro 1.13 ± 0.07 96.9 1.8 ± 0.1 99.3 2.4 ± 0.5 97.8 essential to animal nutrition.
Leu† 1.0700 ± 0.0004 95.1 0.88± 0.05 98.9 1.11 ± 0.03 96.7 Profiles of the amino acid compositions
Thr† 1.60 ± 0 .02 97.8 1.42 ± 0.04 99.4 1.5 ± 0.5 98.0
obtained for coffee pulp, citrus pulp, and dis-
Ala 1.77 ± 0.05 96.1 1.71 ± 0.05 99.2 2.1 ± 0.4 97.2
Lys† 0.18 ± 0.01 98.3 1.07 ± 0.09 99.1 1.00 ± 0.04 96.7
tillery wastes were similar to those reported in
Met† 0.1000 ± 0.0005 98.6 1.14 00 ± 0.0003 99.1 0.8840 ± 0.0002 97.1 the literature (35,36). Other comparisons were
Isl† 0.34 ± 0.04 96.9 0.31 ± 0.01 99.3 0.43 ± 0.03 97.9 not possible due to the unavailability of published
Ser 1.83 ± 0.02 98.4 1.7 ± 0.1 99.5 0.85 ± 0.03 98.1 data. All 17 amino acids investigated were
Val† 0.56 ± 0.02 96.3 0.82 ± 0.03 99.0 0.79 ± 0.04 97.1 detected in the samples. However, no compar-
HO-Pro 0.221± 0.008 97.4 0.0700 ± 0.0008 99.3 0.1700 ± 0.0002 98.1 isons of the quantitative distribution of amino
Tyr 0.36 ± 0.06 96.3 0.460 ± 0.003 99.3 2.2 ± 0.2 98.2 acids in the experimental samples to those
Total 27.2 21.8 28.6 reported in literature were done because reported
EAA % 43.8 55.9 52.8 values were expressed on a constant nitrogen
* Values in table represent the mean of 2 determinations ± standard error of the mean. Abbreviations are as follows:
basis, that is, g/16 g N or as % crude protein,
aspartic acid (Asp); glutamic acid (Glu); phenylalanine (Phe); glycine (Gly); histidine (His); arginine (Arg); proline assuming that all proteins contain exactly 16%
(Pro); leucine (Leu); threonine (Thr); alanine (Ala); lysine (Lys); methionine (Met); isoleucine (Isl); serine (Ser); valine nitrogen.
(Val); hydroxyl-proline (HO-Pro); and tyrosine (Tyr).
† EAA = Essential Amino Acids. In comparing the quantitative distribution of
amino acids among the solid waste samples, it was

677
 Journal of Chromatographic Science, Vol. 47, September 2009

observed that histidine was the major amino acid in all samples, agro-industries could therefore positively impact on Jamaica’s
while methionine and hydroxyl proline were present in the lowest dairy industry.
concentrations. The content of the remaining amino acids were Recoveries of amino acid standards in solid wastes were gen-
variable. In observations made with respect to the quantitative erally good with the exception of orange waste, which had the
distribution of amino acids in liquid wastes, it is important to note lowest overall percentage recovery, ranging from 61.5% (pro-
that, while histidine was not the major amino acid in all liquid line)–94.0% (methionine). In the case of liquid wastes, recov-
wastes, it was present in a relatively high concentration in the eries of amino acids were also good with most being in the high
majority of samples. Methionine and hydroxyl proline were also nineties; continuous still effluent (suspended solids), however,
found to be present in very low concentrations in several liquid had recoveries of amino acids that were mostly in the eighties.
waste samples. The extraction, hydrolysis, and derivatization procedures can
Methionine has been recognized as one of three limiting therefore be described as good, and the results can be used with
amino acids being most likely to be deficient in the rations of confidence. This is highly significant because no published infor-
swine and poultry; the others being lysine and tryptophan (36). mation was found, indicating the use of these methods on mate-
Lysine, methionine, and histidine have also been identified as rials of this nature.
limited in cattle diets (37,38). The low concentrations of methio- No comparisons of the quantitative distribution of amino acids
nine in these samples should not, however, be taken as a negative in the experimental samples were made with the amino acid pro-
indicator because the use of these materials will ultimately be files of two most common animal feeds, corn and soybean, since
decided by their digestibility and how well the amino acids sup- literature values were calculated on a per gram of nitrogen basis,
plied match the nutrient requirements of the animals. which would have included all non-protein nitrogen, thus
High levels of histidine were observed in the samples. resulting in values higher than those in the samples investigated.
However, these results were not in keeping with published data It is important to note, however, that investigations by Elías (35)
on the amino acid profiles of coffee pulp (35), citrus pulp (36), revealed that coffee pulp protein contained similar or higher
and distillery wastes (36), which indicated that although histi- levels of amino acids than soybean flour and corn. Coffee pulp
dine is present, it is not a major constituent of these materials. was also found to be deficient in sulfur-containing amino acids,
The significance of the findings of high levels of histidine in a but had relatively high levels of lysine, as high as those found in
great majority of the samples, should however not be overlooked, soybean meal on a per gram of nitrogen basis.
since it has been recognized that supplying histidine to diets Distillery wastes contain dead yeast cells; hence, observations
deficient in metabolizable histidine can improve protein deposi- with respect to the high levels of total amino acids in some dis-
tion by growing cattle. A determination of the nutritional factors tillery samples were not surprising. Yeast protein has been
that influence the efficiency with which histidine is utilized in described as excellent for a vegetable protein and is about equiv-
cattle could, therefore, improve on current methods of balancing alent in quality to soybean protein. Both are rich in lysine and
diets to meet protein requirements (37). Wastes from these three deficient in sulphur-containing amino acids. In addition,

Table II. Amino Acid Content and Recoveries of Citrus Processing Wastes as Determined by RP-HPLC of PTC-Amino Acids*
Amino Grapefruit Rec. Orange Pulp Rec. Press Liquor Rec. Wash Water Rec.
Acid Pulp mg/100 g (%) (mg/100 g) (%) (mg/L) (%) (mg/L) (%)

Asp 10.7 ± 1.0 99.6 3.4 ± 0.5 77.5 0.120 ± 0.007 94.1 0.2650 ± 0.0003 99.2
Glu 6.9 ± 0.6 97.6 3.0 ± 0.7 68.2 0.36 ± 0.04 93.4 0.7020 ± 0.0003 99.1
Phe† 3.3 ± 0.1 98.3 0.92 ± 0.08 68.1 0.89 ± 0.02 94.9 0.23 ± 0.02 99.1
Gly 1.5 ± 0.3 98.4 0.99 ± 0.01 76.0 0.120 ± 0.008 94.2 0.25 ± 0.01 99.2
His† 25.2 ± 4.6 NR 44.4 ± 1.6 NR 0.24 ± 0.02 82.7 4.0. ± 0.5 98.6
Arg† 2.1 ± 0.4 97.9 2.9 ± 0.1 90.7 0.120 ± 0.009 96.1 0.41 ± 0.03 99.5
Pro 6.4 ± 1.2 NR 11..3 ± 0.4 61.5 0.080 ± 0.002 91.8 1.0 ± 0.1 99.2
Leu† 3.4 ± 0.6 97.4 1.9 ± 0.2 77.0 0.076 ± 0.003 93.3 0.34 ± 0.04 99.0
Thr† 4.6 ± 0.1 95.1 4.3 ± 0.2 80.1 0.17 ± 0.02 94.7 0.580 ± 0.001 99.4
Ala 4.2 ± 0.8 95.2 3.1 ± 0.1 74.5 0.22 ± 0.01 93.1 0.38 ± 0.01 99.0
Lys† 4.9 ± 0.3 97.9 2.2 ± 0.5 85.2 0.181 ± 0.009 93.6 0.31 ± 0.02 99.1
Met† 0.389 ± 0.002 99.0 0.68 ± 0.04 94.0 0.040 ± 0.002 95.1 0.28 ± 0.07 99.3
Isl† 0.87 ± 0.02 97.0 0.8 ± 0.2 84.2 0.0600 ± 0.0009 94.4 0.060 ±0.003 99.3
Ser 2.1 ± 0.2 98.2 1.3 ± 0.5 80.5 0.115 ± 0.007 95.4 0.235 ± 0.003 99.4
Val† 1.3 ± 0.4 95.7 1.6 ± 0.9 76.6 0.13 ± 0.02 94.3 0.32 ± 0.01 99.1
HO-Pro 0.234 ± 0.003 98.0 0.29 ± 0.08 87.2 0.142 ± 0.004 95.5 0.2650 ± 0.0003 99.4
Tyr 0.6 ± 0.1 96.7 0.6 ± 0.1 69.1 0.17 ± 0.03 94.7 0.360 ± 0.001 99.3
Total 78.7 83.7 3.20 10.0
EAA% 58.5 71.3 59.6 65.3
* Values in table represent the mean of 2 determinations ± standard error of the mean. Abbreviations are the same as in Table I.
† EAA = Essential Amino Acids

678
Journal of Chromatographic Science, Vol. 47, September 2009

Table III. Amino Acid Content and Recoveries of Distillery Wastes (Dissolved Solids) as Determined by RP-HPLC of
PTC-Amino Acids*

Amino CS Rec. LPS Rec. HPS Rec. LFB Rec. HFB Rec.
Acid (mg/L) (%) (mg/L) (%) (mg/L) (%) (mg/L) (%) (mg/L) (%)

Asp 3.4 ± 0.8 98.9 6.4 ± 0.4 99.7 10.80 ± 0.04 99.1 3.0 ± 0.2 99.4 6.8 ± 1.6 99.3
Glu 1.42 ± 0.08 98.7 2.0 ± 0.3 99.4 6.13 ± 0.09 99.4 4.5 ± 0.1 99.3 7.69 ± 0.06 99.3
Phe† 0.33 ± 0.01 98.4 1.01 ± 0.05 98.9 2.93 ± 0.02 99.3 2.16 ± 0.08 99.2 3.3 ± 0.1 98.9
Gly 0.59 ± 0.03 98.5 0.65 ± 0.03 99.4 2.03 ± 0.49 99.2 0.94 ± 0.02 99.3 2.3 ± 0.9 99.4
His† 2.6 ± 0.2 96.5 2.5 ± 0.7 99.0 2.700 ± 0.003 97.5 3.7000 ± 0.0005 99.2 2.1 ± 0.2 96.6
Arg† 3.5 ± 0.3 99.4 1.8 ± 0.1 99.7 6.500 ± 0.004 100.0 4.1 ± 0.4 99.8 6.5 ± 0.5 100.0
Pro 0.65 ± 0.04 98.3 0.57 ± 0.02 99.4 3.16 ± 0.02 99.5 2.9 ± 0.3 99.6 1.5 ± 0.4 99.4
Leu† 1.1 ± 0.3 98.5 1.00 ± 0.04 99.0 2.3 ± 0.1 99.1 1.360 ± 0.006 99.0 2.6 ± 0.4 99.0
Thr† 0.83 ± 0.07 98.7 1.6 ± 0.3 99.5 1.6 ± 0.1 99.2 1.330 ± 0.008 99.4 1.7 ± 0.2 99.1
Ala 1.7 ± 0.1 98.2 2.5 ± 0.2 99.4 3.0 ± 0.1 98.6 4.3 ± 0.8 99.2 4.4 ± 0.3 98.3
Lys† 0.45 ± 0.02 99.3 1.38 ± 0.07 98.9 0.9 ± 0.1 99.0 2.5 ± 0.4 99.2 2.1 ± 0.1 99.2
Met† 0.3340 ± 0.0002 98.7 0.392 ± 0.005 99.9 ND‡ ND‡ 0.680 ± 0.005 100.0 ND‡ ND‡
Isl† 0.42 ± 0.02 98.5 0.63 ± 0.01 99.3 1.2 ± 0.5 99.2 0.61 ± 0.01 99.3 1.34 ± 0.04 99.2
Ser 1.07 ± 0.07 98.8 0.82 ± 0.02 99.5 3.4 ± 0.2 99.6 0.57 ± 0.05 99.4 1.8 ± 0.5 99.7
Val† 0.83 ± 0.01 98.3 1.1 ± 0.2 99.0 1.6 ± 0.1 99.0 1.0 ± 0.3 99.0 1.88 ± 0.08 98.9
HO-Pro 0.27 ± 0.02 98.8 0.39 ± 0.03 99.4 1.0 ± 0.2 99.3 0.280 ± 0.002 99.3 0.65 ± 0.02 99.2
Tyr 0.23 ± 0.03 98.9 0.40 ± 0.01 99.4 0.462 ± 0.008 99.3 0.3 ± 0.0 99.3 1.1 ± 0.1 99.2

Total 19.7 25.1 49.7 34.2 47.8

EAA % 52.6 45.3 39.7 51.0 45.0

* Values in table represent the mean of 2 determinations ± standard error of the mean. Abbreviations are the same as in Table I. additional abbreviations: continuous still effluent (CS); light
pot still (LPS); heavy pot still (HPS); light fermentor bottom (LFB); heavy fermentor bottom (HFB)
† EAA = Essential Amino Acids
‡ ND = Not Detected

Table IV. Amino Acid Content and Recoveries of Distillery Wastes (Suspended Solids) as Determined by RP-HPLC of
PTC-Amino Acids*

Amino CS Rec. LPS Rec. HPS Rec. LFB Rec. HFB Rec.
Acid (mg/100 g) (%) (mg/100 g) (%) (mg/100 g) (%) (mg/100 g) (%) (mg/100 g) %

Asp 10.4 ± 0.5 87.0 8.6 ± 1.3 95.1 5.1 ± 0.5 94.0 1.1 ± 0.1 99.0 1.4 ± 0.4 97.8
Glu 16.3 ± 4.4 87.5 15.4 ± 2.1 95.9 10.8 ± 0.5 95.2 2.2 ± 0.4 98.4 2.0 ± 0.2 96.5
Phe† 5.0 ± 0.9 81.8 4.52 ± 0.07 91.1 3.9 ± 0.8 93.5 0.6 ± 0.1 98.0 0.7 ± 0.1 98.5
Gly 5.3 ± 0.1 88.8 3.9 ± 0.3 92.6 3.1 ± 0.2 92.0 0.78 ± 0.07 98.3 0.74 ± 0.08 97.1
His† 15.60 ± 0.05 80.3 19.2 ± 1.8 94.6 11.5 ± 3.9 88.4 2.6 ± 0.4 99.2 2.2 ± 0.4 93.4
Arg† 3.8 ± 0.8 93.8 3.56 ± 0.09 96.5 1.9 ± 0.3 96.2 0.58 ± 0.09 98.9 0.542 ± 0.004 98.2
Pro 2.8 ± 0.1 88.9 4.9 ± 0.5 97.8 2.4 ± 0.1 94.1 0.66 ± 0.01 98.6 0.5 ± 0.2 96.9
Leu† 11.5 ± 1.9 89.0 11.2 ± 0.5 95.9 6.7 ± 0.4 92.4 1.7 ± 0.1 97.4 1.4 ± 0.2 95.2
Thr† 5.8 ± 0.9 91.7 4.85 ± 0.07 96.1 3.5 ± 0.1 95.1 0.74 ± 0.08 98.4 0.82 ± 0.01 98.1
Ala 9.7 ± 0.2 88.7 8.8 ± 0.9 95.0 5.81 ± 0.04 91.9 1.33 ± 0.03 97.4 1.2 ± 0.2 95.8
Lys† 14.4 ± 2.1 84.7 8.0 ± 0.3 90.2 5.5 ± 0.2 96.3 1.2 ± 0.1 97.9 1.6 ± 0.2 97.2
Met† 1.0400 ± 0.0005 84.8 1.01 ± 0.05 93.3 0.65 ± 0.05 95.3 0.16 ± 0.06 97.8 0.18 ± 0.01 98.6
Isl† 5.8 ± 0.2 89.8 6.8 ± 0.2 97.4 4.09 ± 0.01 95.1 0.87 ± 0.01 98.8 0.78 ± 0.02 98.6
Ser 4.5 ± 0.9 91.4 4.1 ± 0.1 96.3 2.4 ± 0.2 95.1 0.62 ± 0.06 98.7 0.61 ± 0.05 97.7
Val† 4.3 ± 0.3 81.6 8.1 ± 0.1 94.8 5.4 ± 0.3 92.7 1.23 ± 0.08 97.2 1.1 ± 0.1 96.1
HO-Pro ND‡ ND‡ ND‡ ND 0.4890 ± 0.0002 95.9 0.0300 ± 0.0007 98.7 ND ND
Tyr 3.6 ± 0.1 90.0 3.6 ± 0.5 96.0 3.3 ± 0.7 95.9 0.61 ± 0.03 98.8 0.61 ± 0.07 98.5
Total 119.8 116.5 76.5 17.0 16.4
EAA% 56.1 57.7 56.3 57.1 56.8

* Values in Table represent the mean of 2 determinations ± standard error of the mean. Abbreviations are the same as in Table I and III.
† EAA = Essential Amino Acids
‡ ND = Not Detected

679
 Journal of Chromatographic Science, Vol. 47, September 2009

because 20% of the crude protein nitrogen in yeast is in the form 12. J. Le Boucher, C. Charret, C. Coudray-Lucas, J. Giboudeau, and L. Cynober.
Amino acid determination in biological fluids by automated ion-exchange chro-
of nucleic acids, there is a limit as to how much can be fed matography: performance of Hitachi L-8500A. Clin. Chem. 43(8): 1421–1428
because excessive nucleic acid intake can result in elevated uric (1997).
13. M.J. González-Castro, J. López-Hernandez, J. Simal-Lozano, and M.J. Oruña-
acid levels in the blood (39). Concha. Determination of amino acids in green beans by derivatization with
phenylisothiocyanate and high-performance liquid chromatography with ultravi-
olet detection. J. Chromatogr. Sci. 35: 181–185 (1997).
14. R.L. Heinrikson and S.C. Meredith. Amino acid analysis by reverse-phase high-
Conclusion performance liquid chromatography: precolumn derivatization with phenylisoth-
iocyanate. Anal. Biochem. 136: 65–74 (1984).
The supply of protein and amino acids in animal diets repre- 15. C.-Y. Yang and F.I. Sepulveda. Separation of phenylthiocarbomyl amino acids by
high-performance liquid chromatography on spherisorb octadecylsilane
sents a significant cost of production. Supplementation of columns. J. Chromatogr. 346: 413–416 (1985).
animal feeds with citrus, coffee, and rum distilleries wastes, 16. V. Fierabracci, P. Masiello, M. Novelli, and E. Bergamini. Application of amino
acid analysis by high-performance liquid chromatography with phenylisothio-
which contain valuable amino acids can no doubt considerably cyanate derivatization to the rapid determination of free amino acids in biological
reduce the cost of imported feeds, especially from the standpoint samples. J. Chromatogr. 570: 285–291 (1991).
17. S.A. Cohen and D.P. Michaud. Synthesis of a fluorescent derivatizing reagent, 6-
of Jamaica. The total amino acid profile is a good indicator of the aminoquinolyl-N-hydroxysuccinimidyl carbamate, and its application for the
potential nutritive value of these agro-industrial wastes. The analysis of hydrolysate amino acids via high-performance liquid chromatography.
Anal. Biochem. 211(2): 279–287 (1993).
results of this study suggest that the HPLC determination and 18. M. Reverter, T. Lundh, and J.E. Lindberg. Determination of free amino acids in pig
quantitation of the phenylthiocarbamyl derivatives of amino plasma by precolumn derivatization with 6–aminoqinolyl-N-hydroxysuccin-
imidyl carbamate and high-performance liquid chromatography. J. Chromatogr.
acids can undoubtedly be used in the development of the total B: Biomed. Sci. Appl. 696(1): 1–8 (1997).
amino acid profiles of agro-industrial wastes. 19. P. Lindroth and K. Mopper. High performance liquid chromatographic determi-
nation of subpicomole amounts of amino acids by precolumn fluorescent deter-
mination with o-phthaldialdehyde. Anal. Chem. 51: 1667–1674 (1979).
20. B.N. Jones and J.P. Gilligan. O-phthaldialdehyde precolumn derivatization and
Acknowledgement reversed-phase high performance liquid chromatography of polypeptide
hydrolysates and physiological fluids. J. Chromatogr. 266: 471–482 (1983).
21. G. Ali Qureshi, L. Fohlin, and J. Bergström. Application of high-performance
This study was financed by grants provided by the liquid chromatography to the determination of free amino acids in physiological
Environmental Foundation of Jamaica (EFJ) and the University fluids. J. Chromatogr. 297: 91–100 (1984).
22. D. Fekkes, A. van Dalen, M. Edelman, and A. Voskuilen. Validation of the deter-
of the West Indies, Mona, Jamaica. The authors would also like to mination of amino acids in plasma by high – performance liquid chromatography
thank, Dr. Lawrence Willliams, who assisted with editing the using automated pre-column derivatization with o-phthaldialdehyde.
J. Chromatogr. B: Biomed. Sci. Appl. 669(2): 177–186 (1995).
manuscript and Mr. Timon Waugh, Dr. Percival Miller and Ms. 23. J.Y. Chang, R. Knecht, and D.G. Braun. Amino acid analysis at the picomole level.
Paulette Tavares who facilitated sample collection and provided Biochem. J. 199: 547–555 (1981).
24. J.Y. Chang, R. Knecht, and D.G. Braun. Methods in Enzymology 91. C.H.W. Hirs,
technical information relevant to the coffee and citrus industries and S. N. Timasheff, Eds. Academic Press, New York, NY, pp. 41–48 (1983).
respectively. 25. Y. Tapuchi, D.E. Schmidt, W. Lindner, and B. L. Karger. Dansylation of amino
acids for high-performance liquid chromatography analysis. Anal. Biochem. 115:
123–129 (1981)
26. N. Kaneda, M. Sato, and K. Yagi. Analysis of dansyl amino acids by reverse phase
References high-performance liquid chromatography. Anal. Biochem. 127(1): 49–54 (1982).
27. S. Einarsson, B. Josefsson, and S. Lagerkvist. Determination of amino acids with
9-fluorenylmethyl chloroformate and reversed-phase high performance liquid
1. R. Gilbert. The world feed industry. In: FAO Animal Production and Health: chromatography. J. Chromatogr. 282: 609–618 (1983).
Protein sources for the animal industry: Expert Consultation and Workshop, 28. S. Einarsson, S. Folestad, B. Josefsson, and S. Lagerkvist. High-resolution reversed
Bangkok, 29 April – 3 March 2002. Food and Agriculture Organization of the phase liquid chromatography system for the analysis of complex solutions of pri-
United Nations, Ed. Rome pp. 1-8 (2004) mary and secondary amino acids. Anal. Chem. 58: 1638–1643 (1986).
2. Food and Agriculture Organization (FAO). Executive Summary. In: FAO Animal 29. H. Kotaniguchi and M. Kawakatsu. Automatic amino acid analysis utilizing 4-
Production and Health: Protein sources for the animal industry: Expert fluoro-7-nitrobenzo-2-oxa-1,3-diazole. J. Chromatogr. 420: 141–145 (1987).
Consultation and Workshop, Bangkok, 29 April – 3 March 2002. Food and 30. C. Aoyama, T. Santa, M. Tsunoda, T. Fukushima, C. Kitada, and
Agriculture Organization of the United Nations, Ed. Rome pp. ix-xxv (2004) K. Imai. A fully automated amino acid analyzer using NBD-F as a fluorescent
3. S. Thomas. Jamaica and Dep: Oilseeds and Products Annual 2004. USDA derivatization reagent. Biomed. Chromatogr. 18(9): 630–636 (2004).
Foreign Agriculture Service Global Agriculture Information Network Report. pp. 31. S.A. Cohen. Amino acid analysis using pre-column derivatization with 6-amino-
1–9 (2004). quinolyl-N-hydroxylsuccinimidyl carbamate: Analysis of hydrolyzed proteins
4. Y. Toride. Lysine and other amino acids for feed: production and contribution to and electroblotted samples. In: Methods in Molecular Biology, 211: Protein
protein utilization in animal feeding. In: FAO Animal Production and Health: Sequencing Protocols, 2nd ed. B.J. Smith, Ed. Humana Press Inc., Totowa, NJ.
Protein sources for the animal industry: Expert Consultation and Workshop, pp. 143–154 (2003).
Bangkok, 29 April – 3 March 2002. Food and Agriculture Organization of the 32. S.J. Watson. The constituents of animal products. In: Feeding of Livestock.
United Nations, Ed. Rome pp. 161–184 (2004). Thomas Nelson and Sons Ltd., New York, pp.16-31 (1949).
5. P. Hušek. Rapid derivatization and gas chromatographic determination of amino 33. P. Hušek and K. Macek. Gas chromatography of amino acids.
acids. J. Chromatogr. 552: 289–299 (1991). J. Chromatogr. 113(2): 139–230 (1975).
6. K.L. Woo and D.K. Chang. Determination of 22 protein amino acids as n (o)-tert.- 34. L.A Maynard and J.K. Loosli. The proteins and their metabolism. In: Animal
butyldimethylsilyl derivatives by gas chromatography. J. Chromatogr. 638: Nutrition. McGraw-Hill Book Company, New York, pp. 115–153 (1969).
97–107 (1993). 35. L.G. Elías. Chemical composition of coffee berry by-products. In: Coffee Pulp-
7. J. Wang, Z.H. Huang, D.A. Gage, and J.T. Watson. Analysis of amino acids by gas Composition, Technology and Utilization. R. Bressani & J.E. Braham (Eds.).
chromatography–flame ionization detection and gas chromatography–mass spec- Ottawa, Ont. IDRC-108e, pp. 11–16 (1979).
trometry: simultaneous derivatization of functional groups by an aqueous-phase 36. C.M. Lyman, K.A. Kuiken, and F. Hale. Amino acids in feedstuffs: Essential amino
chloroformate-mediated reaction. J. Chromatogr. A 663: 71–78 (1994). acid content of farm feeds. J. Agric. Food Chem. 4(12): 1008–1013 (1956).
8. J. Smith. Developments in amino acid analysis using capillary electrophoresis. 37. K.C. McCuistion, E.C. Titgemeyer, M.S. Awawdeh, and D.P. Gnad. Histidine uti-
Electrophoresis 18: 2377–2392 (1997). lization by growing steers is not negatively affected by increased supply of either
9. D.H. Spackman, W.H. Stein and S. Moore. Automatic reading apparatus for use ammonia or amino acids. J. Anim. Sci. 82: 759–769 (2004).
in the chromatography of amino acids. Anal. Chem. 30(1): 190–1205 (1958). 38. R.H. Greenwood and E.C. Titgemeyer. Limiting amino acids for growing Holstein
10. S.E. Moller. Quantification of physiological amino acids by gradient ion- steers limit-fed soybean hull-based diets. J. Anim. Sci. 78: 1997–2004 (2000).
exchange high-performance liquid chromatography. J. Chromatogr. 613: 39. C.W. Stone. Yeast products in the feed industry: A practical guide for feed profes-
223–230 (1993) sionals. Diamond V. Mills, Cedar Rapids, Iowa, pp. 1–16 (1998).
11. M.W. Dong and J.R. Grant. High-speed liquid chromatographic analysis of amino
acids by post-column sodium hypochlorite, o-phthaldehyde reaction. Manuscript received June 23, 2008;
J. Chromatogr. 327: 17–25 (1985). revision received January 19, 2009.

680