Identification and occurrence of the novel alkaloid pentahydroxypentyl-tetrahydro-β-carboline-3-carboxylic acid as a tryptophan glycoconjugate in fruit juices and jams

Article abstract of DOI:10.1021/jf020090m

The novel carbohydrate-derived β-carboline, 1-pentahydroxypentyl-1,2,3,4-tetrahydro-β-carboline-3-carboxylic acid, was identified in fruit- and vegetable- derived products such as juices, jams, and tomato sauces. This compound occurred as two diastereoiso

Full text of DOI:10.1021/jf020090m

4690 J. Agric. Food Chem. 2002, 50, 4690−4695
Identification and Occurrence of the Novel Alkaloid
Pentahydroxypentyl-tetrahydro-â-carboline-3-carboxylic Acid as
a Tryptophan Glycoconjugate in Fruit Juices and Jams
TOMAS HERRAIZ* AND JUAN GALISTEO
Instituto de Fermentaciones Industriales, CSIC, Juan de la Cierva, 3, 28006, Madrid, Spain
The novel carbohydrate-derived â-carboline, 1-pentahydroxypentyl-1,2,3,4-tetrahydro-â-carboline-3-
carboxylic acid, was identified in fruit- and vegetable-derived products such as juices, jams, and
tomato sauces. This compound occurred as two diastereoisomers, a cis isomer (the major compound)
and a trans isomer, ranging from undetectable amounts to 6.5 µg/g. Grape, tomato, pineapple, and
tropical juices exhibited the highest amount of this alkaloid (up to 3.8 mg/L), whereas apple, banana,
and peach juices showed very low or nondetectable levels. This tetrahydro-â-carboline was also
found in jams (up to 0.45 µg/g), and a relative high amount was present in tomato concentrate (6.5
µg/g) and sauce (up to 1.8 µg/g). This â-carboline occurred in fruit-derived products as a
glycoconjugate from a chemical condensation of D-glucose and L-tryptophan that is highly favored at
low pH values and high temperature. Production, processing treatments, and storage of fruit juices
and jams can then release this â-carboline. Fruit-derived products and other foods containing this
compound might be an exogenous dietary source of this glucose-derived tetrahydro-â-carboline.
KEYWORDS: Tetrahydro-â-carbolines; tetrahydro-â-carboline-3-carboxylic acid; alkaloids; tryptophan;
glucose; fruit juices; jams; sauces; Maillard reaction
INTRODUCTION
occurring substances readily produced during fruit production,
processing, and storage. Tetrahydro-â-carbolines have been also
found in many other foodstuffs (17-20). Then, dietary sources
provide tetrahydro-â-carbolines that may subsequently ac-
cumulate in biological tissues and fluids.
1,2,3,4-Tetrahydro-â-carbolines are naturally occurring tri-
cyclic indole derivatives produced from indole-ethylamines and
aldehydes or R-ketoacids through Pictet-Spengler condensation
(1). Similarly, 1,2,3,4-tetrahydro-â-carboline-3-carboxylic acids
arise from a condensation between L-tryptophan and aldehydes.
This latter reaction readily occurs in foods and is temperature-
and pH-dependent (2).
Research in the last two decades has pointed out the occur-
rence of tetrahydro-â-carbolines and â-carbolines in biological
tissues and fluids (3-7). Those compounds might function as
neuromodulators owing to their actions on serotonin uptake,
the benzodiazepine receptor, and monoamine oxidase. They have
also been increasingly studied in relation to alcoholism (6, 8).
Collins and co-workers have reported that N-methylated car-
bolines may act as endogenous neurotoxins (9, 10). 1-Methyl-
1,2,3,4-tetrahydro-â-carboline-3-carboxylic acid, the correspond-
ing tryptophan-acetaldehyde condensation product, is a precursor
of mutagens upon nitrosation (11), and may affect neuronal cell
survival in vitro (12). However, a full delineation of the biolog-
ical activity of this family of compounds is still desirable to
assign a particular biological effect, if any, to each specific
compound.
In the last two years, the possible formation of tryptophan-
derived glycoconjugates, and particularly carbohydrate-derived
â-carbolines, in foods following a reaction of tryptophan and
carbohydrates has become evident (21-23). Thus, Herderich
and co-workers (21) characterized polyol tetrahydro-â-carboline-
3-carboxylic acids and studied N- and C-tryptophan glycocon-
jugates in human urine, whereas Ro¨nner et al. (22) have showed
the formation of these compounds in model reactions of
tryptophan and glucose. However, so far, the determination and
specific identification of these novel polyol tetrahydro-â-
carbolines in foods remained to be accomplished. The presence
of the carbohydrate-derived â-carboline, 1-pentahydroxypentyl-
tetrahydro-â-carboline-3-carboxylic acid as two diastereoiso-
mers, in fruit-derived products, and its quantitative determination
in fruit juices and jams has now been established in this study.
In addition, the possible chemical formation of this compound
is evaluated in terms of the influence of pH and temperature.
MATERIALS AND METHODS
In the past few years, we have reported the presence of
tetrahydro-â-carbolines in fruit and fruit-derived products such
as juices and jams (13-16). These compounds are naturally
Reference Compounds and Samples. The synthesis of carbohydrate-
derived tetrahydro-â-carbolines was achieved by a Pictet-Spengler type
reaction. 1-(1,2,3,4,5-Pentahydroxypent-1-yl)-1,2,3,4-tetrahydro-â-car-
boline-3-carboxylic acid was synthesized from L-tryptophan and
D-glucose according to reference 22. Briefly, L-tryptophanate sodium
* Corresponding author. Tel: 34-915622900. Fax: 34-915644853.
E-mail: therraiz@ifi.csic.es.
10.1021/jf020090m CCC: $22.00 © 2002 American Chemical Society
Published on Web 06/27/2002

Tetrahydro-â-Carbolines in Fruit Juices
J. Agric. Food Chem., Vol. 50, No. 16, 2002 4691
Table 1. Concentration (mg/L) of 1,2,3,4,5-Pentahydroxypentyl-1,2,3,4-tetrahydro-â-carboline-3-carboxylic Acid Alkaloids (cis and trans isomers) in
Commercial Fruit and Vegetable Juicesa
trans isomer
SD
cis isomer
SD
juice
orange
pineapple
grape
pineapple + grape
peach nectar
peach + grape
tomato
banana nectar
apple
tropical
vegetable
multifruit
n
X
range
X
range
12
7
3
3
2
4
7
2
3
3
1
2
0.037
0.17
0.59
0.22
-
0.21
0.24
-
0.012
0.2
0.23
0.055
-
0.09
0.14
-
0.02−0.063
0.0−0.48
0.43−0.85
0.16−0.27
-
0.11−0.31
0.06−0.43
-
0.16
0.60
2.18
0.71
-
0.75
0.99
-
0.055
0.67
0.76
0.16
-
0.3
0.57
-
0.1−0.28
0.036−1.7
1.5−3.0
0.57−0.89
-
0.39−1.12
0.27−1.89
-
-
-
0.2
0.04
0.08
0.15
-
0.05
0.05−0.35
-
0.045−0.11
0.64
0.25
0.30
0.43
-
0.15
0.20−1.07
-
0.19−0.40
a Most of the juices were from concentrate, but some from orange, pineapple, and tomato were from direct fruit (not from concentrate) and others were refrigerated
juices. Not detectable or below detection limit (−).
Table 2. Concentration (µg/g) of
salt was formed from L-tryptophan (1.5 mmol) and NaOH (1.5 mmol)
in methanol, and evaporated to dryness. After addition of D-glucose
(7.5 mmol) in methanol the reaction was kept under reflux for 2.5 h.
The solution was concentrated under vacuum to ca. 10 mL, added
dropwise to methanolic 2 N HCl (10 mL) at 0-5 °C, and stirred for 1
h. The mixture was concentrated to dryness. Isolation of glucose-derived
tetrahydro-â-carboline was first carried out on a Dowex resin 50W ×
8 (15 g) (Fluka) placed on a sintered disk filter funnel, which was eluted
with 0.2 M ammonia and 0.2 M ammonia/acetonitrile (1:1) solutions
until no further product was detectable by HPLC, and subsequently
dried under vacuum. After redissolving in water, a further isolation
and purification of the compounds was carried out on a 50 mm × 20
mm i.d. chromatographic column packed with C-18 sorbent by eluting
with an increasing gradient of acetonitrile. The fractions containing
two compounds, corresponding to the major cis and minor trans
diastereoisomers, with the same fluorescence and UV spectra by
RPHPLC, were dried under vacuum to give glucose-derived tetrahydro-
â-carboline-3-carboxylic acid. The major compound (cis isomer),
(1R,3S)-1-(1,2,3,4,5-pentahydroxypentyl)-1,2,3,4-tetrahydro-â-carboline-
3-carboxylic acid, was characterized by 1H- and 13C-NMR (Varian XL-
400 spectrometer) giving signals in agreement with those of previous
results (21, 22). 1H NMR (CD3OD/TFA): δ 3.34 (m, 1H, H-4a), 3.57
(dd, 1H, H-4b), 3.96 (dd, 1H H-4′), 3.96-4.07 (m, 2H, H-5′), 4.13 (d,
1H, H-3′), 4.48 (m, 1H, H-2′), 4.52 (dd, 1H, J (3,4a) ) 5 Hz, J (3,4b)
) 12 Hz, H-3), 4.82 (m, 1H, J (1′,1) ) 1.7 Hz, J (1′, 2′) ) 4 Hz,
H-1′), 5.24 (s, 1H, H-1), 7.25 (t, 1H, J ) 8 Hz, H-6), 7.35 (t, 1H, J )
8 Hz, H-7), 7.56 (d, 1H, J ) 8 Hz, H-8) 7.66 (d, 1H, J ) 8 Hz, H-5).
13C NMR (CD3OD/TFA): δ 22.8 (C-4), 56.9 (C-3), 57.6 (C-1), 64.5
(C-5′), 71.4 (C-1′), 71.9 (C-2′), 72.2 (C-3′), 72.8 (C-4′), 108.6 (C-4a),
112.5 (C-8), 118.9 (C-5), 120.7 (C-6), 123.6 (C-7), 127.4 (C-4b), 128.6
(C-9a), 138.6 (C-8a), 171 (COOH). MS (ESI-positive mode) [M +
H]+ m/z 367; MS (ESI-negative mode) [M - H]- m/z 365, 303, 215.
Analysis by RP(C18)-HPLC-MS (ESI-positive mode) (see below for
chromatographic conditions) confirmed that pentahydroxypentyl-1,2,3,4-
tetrahydro-â-carboline-3-carboxylic acid occurs as two diastereoisomers
with the same mass spectra eluting at 13.5 min (minor compound) and
15.8 min (major compound cis isomer). MS (fragmentation voltage 50
V): m/z 367 [M + H]+, and MS (fragmentation voltage 120 V): m/z
247 [M + H - C4H8O4]+ (100), 230 (96), 367 [M + H]+ (82), 350
(49), 276 (48), 332 (45), and 294 (15). Alternatively, cis-(1R,3S)-1-
(1,2-dihydroxyethyl)-1,2,3,4-tetrahydro-â-carboline-3-carboxylic acid
was synthesized by condensation of L-tryptophan and D-glyceraldehyde
in 0.05 N sulfuric acid providing a solid that was filtered off, and its
structure was confirmed by NMR and MS: m/z 277 [M + H]+, 204
[M + H-73]+. This compound was chosen as an internal standard (IS)
for quantitative purposes owing to its structure and good chromato-
graphic properties, and because it afforded recovery percentages similar
to those of pentahydroxypentyl tetrahydro-â-carboline-3-carboxylic acid
during the isolation procedure.
1,2,3,4,5-Pentahydroxypentyl-1,2,3,4-tetrahydro-â-carboline-3-carboxylic
Acid Alkaloids (cis and trans isomers) in Commercial Jams and
Tomato Sauces
trans isomer
cis isomer
n
X
X
strawberry
pineapple
3
1
1
1
2
2
1
4
1
0.08 ± 0.014
0.089
0.31 ± 0.02
0.19
0.168
0.37
apricot
0.044
prune
peach
bitter orange
raspberry
tomato sauce/ketchup
tomato concentrate
0.079
a
-
-
0.019 ± 0.01
0.07
0.115 ± 0.002
0.21
1.21 ± 0.62
5.19
0.35 ± 0.16
1.29
a Not detectable or below detection limit (−).
(not refrigerated), nectars and squeezed (fresh and refrigerated, not from
concentrate) fruit juices, and also commercial jams, tomato sauces, and
ketchup (Tables 1 and 2), were purchased in local supermarkets and
used for analysis of pentahydroxypentyl-1,2,3,4-tetrahydro-â-carboline-
3-carboxylic acid.
Isolation of Pentahydroxypentyl-tetrahydro-â-carboline-3-car-
boxylic Acid. Pentahydroxypentyl-tetrahydro-â-carboline-3-carboxylic
acid was isolated using SCX-solid-phase extraction procedures as
follows for the differenct food products. (A) Fruit juice (20 mL) was
centrifuged (15000g, 0 °C) for 10-15 min. An aliquot of supernatant
(5 mL) was spiked with 1 mL of cis-(1R,3S)-1-(1,2-dihydroxyethyl)-
1,2,3,4-tetrahydro-â-carboline-3-carboxylic acid solution (5 mg/L) used
as internal standard (IS), 0.5 mL of semicarbazide (Sigma) solution 10
mg/mL, and 1 mL of 0.1 M HCl, and slowly passed through Bond
Elut 500 mg/3 mL benzenesulfonic acid SCX columns (Varian, Harbor
City, CA) using a vacuum manifold. (B) Jams, concentrates, and sauces
(5 g) were homogenized using an UltraTurrax homogenizer with 10
mL of 0.6 M HClO4 containing 1 mg/mL semicarbazide and centrifuged
(15000g, 10-15 min, 0-5 °C). An aliquot of supernatant (5.5 mL),
was added with 1 mL of 1-(1,2-dihydroxyethyl)-1,2,3,4-tetrahydro-â-
carboline-3-carboxylic acid solution (5 mg/L) and 1 mL of HCl 0.1
M, and loaded onto an SCX column. SCX columns were washed with
6 mL of 0.6 M HCl, 2 mL of methanol, and 6 mL of Milli-Q purified
water. Then, elution of pentahydroxypentyl tetrahydro-â-carbolines was
carried out with 4 mL of 0.4 M phosphate buffer pH 9.2, and injected
onto RPHPLC.
Chromatographic Analysis. Chromatographic analysis of pentahy-
droxypentyl-tetrahydro-â-carboline was performed by RPHPLC and
fluorescence detection. A 150 mm × 3.9 mm, 5 µm, Nova-pak C18
column (Waters, Milford, MA) was used for separation. Chromato-
graphic conditions were as follows: 50 mM ammonium phosphate
buffer (pH 3) (buffer A) and 20% of A in acetonitrile (buffer B),
gradient programmed from 0% (100% A) to 5% B in 16 min. The
Different commercial samples of packed fruit and vegetable juices,
in both bottles and cartons, including juices made from concentrate

4692 J. Agric. Food Chem., Vol. 50, No. 16, 2002
Herraiz and Galisteo
Figure 1. RPHPLC chromatograms of fruit juices: grape juice (a), pineapple juice (b), and tomato juice (c), and pentahydroxypentyl-1,2,3,4-tetrahydro-
â-carboline-3-carboxylic acid reference compounds (cis and trans isomers) (d). IS: cis-1-(1,2-dihydroxyethyl)-1,2,3,4-tetrahydro-â-carboline-3-carboxylic
acid.
averaged 92% (n ) 4, grape juice), whereas the reproducibility was
RSD 1.8% (n ) 4, grape juice). The detection limit was below 0.01
mg/L. Blanks and control samples (100 mg/L L-tryptophan, 100 mg/L
glucose) did not give artifacts during isolation and analysis. Confirma-
tion of the identity of isolated pentahydroxypentyl tetrahydro-â-
carboline was established by HPLC retention times, fluorescence
spectra, and coelution with authentic standards. Thus, fluorescence
spectra of the HPLC peaks were compared with those of reference
compounds. For this, eluting peaks corresponding to pentahydroxypentyl
tetrahydro-â-carboline in real samples and standards were trapped into
the flow cell of the fluorescence detector by stopping the solvent pump,
and excitation and emission spectra were monitored. In addition, several
samples of juices and tomato concentrate were subjected to HPLC-
MS as reported below.
Chemical Identification by HPLC-MS. Samples of fruit juices
and tomato concentrate were subjected to SCX-extraction as described
above. The eluting fractions corresponding to phosphate buffer pH 9.2
were concentrated under a nitrogen stream and analyzed by HPLC-
MS. Chemical identification was accomplished by HPLC-MS on a
3.9 mm × 150 mm Novapak C18 column (Waters), by using an
HPLC-MSD series 1100 (Hewlett-Packard) (electrospray-positive ion
mode). Eluents consisted of A, formic acid (0.5%); and B, 0.5% formic
acid in acetonitrile. The gradient was 0% B to 12.5% B in 25 min.
Flow was 0.7 mL/min; cone voltage was 50 or 120 V for higher
fragmentation; capillary voltage was 4000 V; drying gas temperature
was 350 °C; gas flow was 10 L/min; and the mass range was 50-
1000 amu.
Figure 2. Emission fluorescence (excitation at 270 nm) of pentahydroxy-
pentyl-1,2,3,4-tetrahydro-â-carboline-3-carboxylic acid (cis and trans
isomers) in fruit juices and standard. Fluorescence data are normalized
to 100%.
Chemical Factors Affecting the Formation of Pentahydroxypen-
tyl Tetrahydro-â-Carboline. According to previous results, reaction
to give tetrahydro-â-carbolines is highly affected by chemical factors
such as pH and temperature (2). To evaluate this effect on the possible
formation of glucose-derived tetrahydro-â-carbolines, several test tubes
containing L-tryptophan (0.5 g/L) and D-glucose (1 g/L) were incubated
in duplicate at different pH values (pH 1 from 0.8% v/v HCl conc +
2.2 g/L NaCl, and 50 mM phosphate buffer solutions of pH values 3,
flow rate was 1 mL/min, the column temperature was 40 °C, and the
injection volume was 20 µL. Fluorescent detection was set at 270 nm
for excitation and 343 nm for emission.
Quantitation was obtained from calibration curves constructed from
5, 7, and 9), and temperatures (25, 37, and 70 °C) while following the
known standard solutions of pentahydroxypentyl-tetrahydro-â-carboline-
formation of this carboline by RPHPLC. In addition, whole grapes were
crushed in the laboratory and a freshly prepared grape juice was
submitted to heating (refluxing for up to 8 h), and subsequently analyzed
for pentahydroxypentyl-tetrahydro-â-carboline-3-carboxylic acid.
3-carboxylic acid reference compound (from 0.05 to 2 mg/L), spiked
with dihydroxyethyl-tetrahydro-â-carboline-3-carboxylic acid as IS, and
analyzed through the entire procedure as above. The same response
factor was used for the two isomers. The recovery of the compound

Tetrahydro-â-Carbolines in Fruit Juices
J. Agric. Food Chem., Vol. 50, No. 16, 2002 4693
Figure 3. HPLC−MS analysis (electrospray-positive mode) and mass spectra of pentahydroxypentyl-1,2,3,4-tetrahydro-â-carboline-3-carboxylic acid.
Fragmentation voltage: 50 V (a) and 120 V (b).
Figure 4. HPLC−MS analysis and reconstructed ion chromatogram at m/z 367 [M + H]+ of grape juice (a), pineapple juice (b), and tomato juice (c).
RESULTS AND DISCUSSION
emission maximum around 345 nm when excited at 270 nm
(Figure 2). A further unequivocal proof of the existence of this
compound in those products was obtained by HPLC-MS
(Figure 3). Extracted samples of fruit and vegetable juices and
tomato concentrate gave mass spectra corresponding to pen-
tahydroxypentyl tetrahydro-â-carboline-3-carboxylic acid (Fig-
ure 4). The mass spectra showed the presence of two isomers
of this carboline with protonated molecular ions [M + H]+ at
m/z 367 that coeluted with the corresponding reference standards
under the same conditions. As occurs for other 1,3-disubstituted
tetrahydro-â-carbolines detected in fruit juices (14), the novel
polyol tetrahydro-â-carboline appeared as two diastereoisomers
corresponding to 1R,3S (cis) and 1S,3S (trans) configurations
(Figure 5). The trans isomer (1S,3S) eluted as the first peak in
RPHPLC and the cis isomer (1R,3S) eluted as the second peak
(21). The cis isomer (second chromatographic peak) was the
major compound in fruit- and vegetable-derived samples, and
the ratio cis/trans generally ranged from 3 to 4.5.
In previous studies we have reported the chemical identifica-
tion and quantitative analysis of tetrahydro-â-carbolines in fruits
and fruit-derived products such as juices and jams (14, 15).
During that research we had detected possible unknown
tetrahydro-â-carbolines in fruit-derived samples and we have
now tried to isolate and identify these possible new compounds.
Thus, SCX-extracted fruit-derived products (i.e., juices, jams,
tomato sauces, and concentrates) gave chromatographic peaks
coeluting with authentic synthetic standards of pentahydroxy-
pentyl-tetrahydro-â-carboline-3-carboxylic acid (cis and trans
diastereoisomers) (Figure 1). Those HPLC peaks coeluting with
standards of this carboline were trapped into the flow cell of
the fluorescence detector, and the fluorescence spectra were
monitored. Excitation and emission profiles from diastereoiso-
meric peaks detected in fruit-derived products were very
consistent with those from authentic standards exhibiting an

4694 J. Agric. Food Chem., Vol. 50, No. 16, 2002
Herraiz and Galisteo
tetrahydro-â-carboline (Figure 5). As shown in Figure 6, the
chemical formation of this glucose-derived tetrahydro-â-car-
boline increased exponentially at low pH and high temperature.
This behavior agrees with previous results for other tetrahydro-
â-carbolines such as 1,2,3,4-tetrahydro-â-carboline-3-carboxylic
acid and 1-methyl-1,2,3,4-tetrahydro-â-carboline-3-carboxylic
acid (2). No reaction to give the polyol tetrahydro-â-carboline
took place at the highest pH values (7 and 9) even at high
temperature, whereas the formation occurred very slowly at low
temperature even in very acidic pH. Pentahydroxypentyl-
tetrahydro-â-carboline-3-carboxylic acid needed much more
extreme conditions to form readily (i.e., low pH and high
temperature) in comparison with the relatively mild conditions
needed for formation of other tetrahydro-â-carboline-3-car-
boxylic acids (2, 13, 14). Considering these results, it should
be expected that food samples containing high levels of glucose
and tryptophan, that are processed at low pH and high
temperature (as occurs in fruit juices and jams) will provide
glucose-derived tetrahydro-â-carboline-3-carboxylic acid in
significant amounts. Figure 7 confirms this assumption for a
freshly obtained grape juice. The concentration of pentahy-
droxypentyl tetrahydro-â-carboline greatly increased during the
heating time. The same is expected to occur during processing,
concentration, and conservation treatments of fruit-derived
products. Thus, differences among samples might reflect distinct
concentration of reactants (glucose and tryptophan) and/or
processing conditions.
The above results reveal for the first time the occurrence of
the carbohydrate-derived tetrahydro-â-carboline pentahydroxy-
pentyl-1,2,3,4-tetrahydro-â-carboline-3-carboxylic acid, in fruit
and vegetable juices, jams, and tomato concentrate and sauces.
This finding extends previous reports on the presence of other
tetrahydro-â-carbolines in juices and jams following a Pictet-
Spengler condensation between tryptophan and fruit aldehydes
(13-16). It is well recognized that tryptophan may undergo
complex transformations giving rise to possible bioactive
compounds (24). Occurrence of tetrahydro-â-carbolines in
commonly ingested foods such as fruit-derived products and
others, strongly evidences an exogenous intake of these
compounds during food and drink consumption. This may
influence the endogenous presence of â-carbolines in tissues,
Figure 5. Pictet−Spengler type condensation of L-tryptophan and D-glucose
to give pentahydroxypentyl-1,2,3,4-tetrahydro-â-carboline-3-carboxylic acid
as two diastereoisomers (cis and trans configuration).
The occurrence of these carbohydrate-derived tetrahydro-â-
carbolines (trans and cis diastereoisomers) in juices is sum-
marized in Table 1. Its content varied both between individual
juices and even more among groups of juices. Grape juices or
mixed fruit juices containing concentrate grape juice exhibited
the highest level of pentahydroxypentyl-tetrahydro-â-carboline-
3-carboxylic acid (up to 3.8 mg/L). Relatively high concentra-
tions of this carboline were also found in tomato and pineapple,
as well as in multifruit and tropical fruit juices. In contrast, it
was not detected in apple juices or banana and peach nectars,
whereas a low amount was detected in orange juices. Most of
the fruit juices analyzed here were made from concentrate juice,
so that this polyol-tetrahydro-â-carboline might have formed
during the processing conditions carried out for juice concentra-
tion (i.e., relatively high temperature and concentration of
reactants). However, a few commercial juices marked as fresh
and refrigerated, and made from direct squeezing of oranges,
pineapples and tomatoes, also seemed to contain this compound,
although in lower amounts. As shown in Table 2, the pentahy-
droxypentyl tetrahydro-â-carboline was also found in com-
mercial jams and tomato sauces. Most of the jams analyzed,
with the exception of those from peach, seemed to contain a
low amount of this tetrahydro-â-carboline reaching up to 0.45
µg/g. Tomato concentrates, sauces, and ketchup contained even
more of this compound, reaching up to 6.5 µg/g. This polyol-
tetrahydro-â-carboline might have formed during the processing
conditions carried out for tomato juice concentration and sauce
production.
This alkaloid comes from a condensation reaction involving
L-tryptophan and D-glucose following by cyclization to give the
Figure 6. Formation of pentahydroxypentyl-1,2,3,4-tetrahydro-â-carboline-3-carboxylic acid (mg/L) from a reaction of D-glucose (1 g/L) and L-tryptophan
(0.5 g/L) at different pH values and temperatures (14 days of reaction, a−c).

Tetrahydro-â-Carbolines in Fruit Juices
J. Agric. Food Chem., Vol. 50, No. 16, 2002 4695
bolines in brain particulate fractions. J. Neurochem. 1992, 59,
511-518.
(11) Wakabayashi, K.; Ochiai, M.; Saito, H.; Tsuda, M.; Suwa, Y.;
Nagao, M.; Sugimura, T. Presence of 1-methyl-1,2,3,4-tetrahy-
dro-â-carboline-3-carboxylic acid, a precursor of a mutagenic
nitroso compound, in soy sauce. Proc. Natl Acad. Sci. U.S.A.
1983, 80, 2912-2916.
(12) Brenneman, D. E.; Page, S. W.; Schultzberg, M.; Thomas, F.
S.; Zelazowski, P.; Burnet, P.; Avidor, R.; Sternberg, E. M. A
decomposition product of a contaminant implicated in L-
tryptophan eosinophilia myalgia syndrome affects spinal cord
neuronal cell death and survival through stereospecific, matura-
tion and partly interleukin-1-dependent mechanisms. J. Phar-
macol. Exp. Ther. 1993, 266, 1029-1035.
(13) Herraiz, T. Occurrence of tetrahydro-â-carboline-3-carboxylic
acids in commercial foodstuffs. J. Agric Food Chem. 1996, 44,
3057-3065.
Figure 7. Formation of pentahydroxypentyl-1,2,3,4-tetrahydro-â-carboline-
3-carboxylic acid (cis and trans isomers) (mg/L) in freshly prepared grape
juice when heated at reflux.
organs, and fluids (2-6) because accumulation in vivo could
arise, at least in part, from food-ingested tetrahydro-â-carbolines
(25). Indeed, pentahydroxypentyl-tetrahydro-â-carboline-3-car-
boxylic acid and other tryptophan glycoconjugates had been
previously detected in human urine (21, 26). In this regard, this
alkaloid might form either endogenously or be absorbed from
the diet. As chemical formation under physiological conditions
(pH 7, 37 °C) is unlikely (Figure 6), it should be assumed that
any possible presence of this polyol tetrahydro-â-carboline in
vivo should surely arise from dietary sources. This is also
supported by the extensive presence of this compound in fruit-
derived products and probably in other foodstuffs. Although
many studies have considered the possible bioactivity of
tetrahydro-â-carbolines and â-carbolines, a full delineation of
the biological activity of this class of compounds remains to
be accomplished.
(14) Herraiz, T. Occurrence of 1,2,3,4-tetrahydro-â-carboline-3-
carboxylic acid and 1-methyl-1,2,3,4-tetrahydro-â-carboline-3-
carboxylic acid in fruit juices, purees, and jams. J. Agric. Food
Chem. 1998, 46, 3484-3490.
(15) Herraiz, T. 1-Methyl-1,2,3,4-tetrahydro-â-carboline-3-carboxylic
acid and 1,2,3,4-tetrahydro-â-carboline-3-carboxylic acid in
fruits. J. Agric. Food Chem. 1999, 47, 4883-4887.
(16) Herraiz, T.; Sanchez, F. Presence of tetrahydro-â-carboline-3-
carboxylic acids in foods by gas chromatography-mass spec-
trometry as their N-methoxycarbonyl methyl ester derivatives.
J. Chromatogr. A 1997, 765, 265-277.
(17) Herraiz, T.; Huang, Z.; Ough, C. S. 1,2,3,4-Tetrahydro-â-
carboline-3-carboxylic acid and 1-methyl-1,2,3,4-tetrahydro-â-
carboline-3-carboxylic acid in wines. J. Agric. Food Chem. 1993,
41, 455-459.
(18) Sen, N. P.; Seaman, S. W.; Lau, B. P. Y.; Weber, D.; Lewis, D.
Determination and occurrence of various tetrahydro-â-carboline-
3-carboxylic acids and the corresponding N-nitroso compounds
in foods and alcoholic beverages. Food Chem. 1995, 54, 327-
337.
(19) Herraiz, T. Analysis of the bioactive alkaloids tetrahydro-â-
carboline and â-carboline in food. J. Chromatogr. A 2000, 881,
483-499.
(20) Gutsche, B.; Herderich, M. HPLC-MS/MS identification of
tryptophan-derived tetrahydro-â-carboline derivatives in food.
Fresenius’ J. Anal. Chem. 1998, 360, 836-839.
(21) Gutsche, B.; Grun, C.; Scheutzow, D.; Herderich, M. Tryptophan
glycoconjugates in food and human urine. Biochem. J. 1999,
343, 11-19.
(22) Ro¨nner, B.; Lerche, H.; Bergmu¨ller, W.; Freilinger, C.; Severin,
T.; Pischetsrieder, M. Formation of tetrahydro-â-carbolines and
â-carbolines during the reaction of L-tryptophan and D-glucose.
J. Agric. Food Chem. 2000, 48, 2111-2116.
ACKNOWLEDGMENT
We thank Carmen Fernandez and Maribel Jimenez for assistance
in sample processing and mass spectrometry, respectively.
LITERATURE CITED
(1) Whaley, W. M.; Govindachari, T. R. The Pictet-Spengler
synthesis of tetrahydroisoquinolines and related compounds. Org.
React. (Tartu) 1951, VI, 151-190.
(2) Herraiz, T.; Ough, C. S. Chemical and technological factors
determining tetrahydro-â-carboline-3-carboxylic acid content in
fermented alcoholic beverages. J. Agric. Food Chem. 1993, 41,
959-964.
(3) Buckholtz, N. S. Neurobiology of tetrahydro-â-carbolines. Life
Sci. 1980, 27, 893-903.
(4) Airaksinen, M. M.; Kari, I. â-carbolines, psychoactive com-
pounds in the mammalian body. Med. Biol. 1981, 59, 21-34.
(5) Melchior, C.; Collins, M. A. The route and significance of
endogenous synthesis of alkaloids in animals. CRC Crit. ReV.
Toxicol. 1982, 10, 313-356.
(6) Myers, R. D. Isoquinolines, beta-carbolines and alcohol drink-
ing: Involvement of opioid and dopaminergic mechanisms.
Experientia 1989, 45, 436-443.
(7) Rommelspacher, H.; May, T.; Susilo, R. â-Carbolines and
Tetrahydroisoquinolines: detection and function in mammals.
Planta Med. 1991, 57 (Suppl.), S85-S92.
(8) Tuomisto, L.; Airaksinen, M. M.; Peura, P.; Eriksson, C. J. P.
Alcohol drinking in the rat: increases following intracerebroven-
tricular treatment with tetrahydro-â-carbolines. Pharmacol. Bio-
chem. BehaV. 1982, 17, 831-836.
(23) Wang, M.; Jin, Y.; Li, J.; Ho, C. Two novel â-carboline
compounds from the Maillard reaction between xylose and
tryptophan. J. Agric. Food Chem. 1999, 47, 48-50.
(24) Friedman, M.; Cuq, J. L. Chemistry, analysis, nutritional value,
and toxicology of tryptophan in food. A review. J. Agric. Food
Chem. 1988, 36, 1079-1093.
(25) Ogawa, Y.; Adachi, J.; Tatsuno, Y. Accumulation of 1-methyl-
tetrahydro-â-carboline-3-carboxylic acid in blood and organs of
rat. A possible causative substance of eosinophilia-myalgia
syndrome associated with ingestion of L-tryptophan. Arch.
Toxicol. 1993, 67, 290-293.
(26) Horiuchi, K.; Yonekawa, O.; Iwahara, K.; Kanno, T.; Kurihara,
T.; Fujise, Y. A hydrophilic tetrahydro-â-carboline in human
urine. J. Biochem. 1994, 115, 362-366.
(9) Collins, M. A.; Neafsey, E. J. â-carboline analogues of N-methyl-
4-phenyl-1,2,3,6-tetrahydropyridine (MPTP): endogenous factors
underlying idiophatic Parkinsonism? Neurosci. Lett. 1985, 55,
179-183.
(10) Matsubara, K.; Neafsey, E. J.; Collins, M. A. Novel S-
Adenosylmethionine-dependent indole-N-methylation of â-car-
Received for review January 25, 2002. Revised manuscript received
May 8, 2002. Accepted May 8, 2002. This work was supported by the
research projects AGL2000-1480 and AGL2000-1569 (CICYT, Spain).
JF020090M