Varietal Differences in the Total and Enantiomeric Composition of Theanine in Tea

Article abstract of DOI:10.1021/jf960432m

Theanine is the main amino acid component in tea. It usually constitutes between 1 and 2% of the dry weight of the tea leaves. It is as prevalent in tea as all other free amino acids combined. Both enantiomers of theanine were found to have a similar sweet taste, with little or no aftertaste. It was found that black and half-green teas (except for Formosa Oolong) contained as much, or more, theanine as green teas. No correlation was found between the absolute concentration of theanine in tea and its enantiomeric composition. An inverse correlation was found between certain grades of tea (e.g., pekoe, Flowery Orange Pekoe, etc.) and the percent of D-theanine present. This could provide the basis for a reproducible, scientific method to grade and/or evaluate teas. Theanine slowly racemizes in aqueous solution. It also undergoes hydrolysis, particularly at basic pH values. By monitoring these processes, information may be gleaned on the production, storage, handling, and shipping of tea and tea products.

Full text of DOI:10.1021/jf960432m

J. Agric. Food Chem. 1997, 45, 353
−363
353
Va r ieta l Differ en ces in th e Tota l a n d En a n tiom er ic Com p osition of
Th ea n in e in Tea
K. Helen Ekborg-Ott, Andre Taylor, and Daniel W. Armstrong*
Department of Chemistry, University of MissourisRolla, Rolla, Missouri 65401
Theanine is the main amino acid component in tea. It usually constitutes between 1 and 2% of the
dry weight of the tea leaves. It is as prevalent in tea as all other free amino acids combined. Both
enantiomers of theanine were found to have a similar sweet taste, with little or no aftertaste. It
was found that black and half-green teas (except for Formosa Oolong) contained as much, or more,
theanine as green teas. No correlation was found between the absolute concentration of theanine
in tea and its enantiomeric composition. An inverse correlation was found between certain grades
of tea (e.g., pekoe, Flowery Orange Pekoe, etc.) and the percent of D-theanine present. This could
provide the basis for a reproducible, scientific method to grade and/or evaluate teas. Theanine
slowly racemizes in aqueous solution. It also undergoes hydrolysis, particularly at basic pH values.
By monitoring these processes, information may be gleaned on the production, storage, handling,
and shipping of tea and tea products.
Keyw or d s: Racemization; beverage; D-amino acid; tea grades; cyclodextrin column; column
switching
INTRODUCTION
Tea is the most widely consumed beverage in the
world. The three main types of tea are (1) green
(
nonfermented) tea, (2) half-green or Oolong (semifer-
mented) tea, and (3) black (fermented) tea. The term
fermented” is something of a misnomer insofar as teas
“
are concerned. The term “oxidation” is more fitting,
since tea fermentation does not require microorganisms
(although they may be present in some green and half-
green teas). The fermentation step consists of the
oxidation of two or more tea flavanols (polyphenolic
compounds) belonging to the catechin group (Wickre-
masinghe, 1978; Finger et al., 1992; Roberts, 1952;
Harler, 1963). This oxidation step is precluded in green
tea, while Oolong tea is partially oxidized, and black
tea is extensively oxidized (Graham, 1992). Green tea
flavor has four characteristic taste elements: bitterness,
astringency, sweetness, and umami (a brothy or savory
taste) (Kawamura and Kare, 1987; Teranishi, 1989;
Nakagawa, 1975). The brothy, sweet, umami taste is
due to amino acids, especially theanine. Theanine, also
known as glutamic acid γ-ethylamide or 5-N-ethyl-
glutamine, exists only in the free (non-protein) form and
is by far the major free amino acid in tea (Selvendran,
F igu r e 1. Reaction of pyroglutamic acid with ethylamine to
form theanine.
The precursors for theanine biosynthesis are glutamic
acid and ethylamine. Ethylamine is a constituent of
many plant roots (Sasaoka et al., 1965; Gray, 1963;
Tunmann and Linde, 1958; Serenkov, 1959) and is
produced by the enzymatic decarboxylation of alanine
(Crocomo and Fowden, 1970; Takeo, 1974). Biosynthe-
sis of theanine occurs in the root of the tea plant with
the aid of the enzyme theanine synthetase (also called
L-glutamic acid ethylamine ligase or L-glutamate ethyl-
amine ligase) (Feldheim et al., 1986; Sasaoka and Kito,
1
964; Sasaoka et al., 1963, 1965; Wickremasinghe and
Perera, 1972; Orihara and Furuya, 1990; Matsuura and
Kakuda, 1990; Sasaoka, 1965; Perera and Wickremas-
inghe, 1971; Chakraborty et al., 1978; Konishi, 1970;
Konishi et al., 1969; Konishi and Kasai, 1968a,b). The
theanine is then translocated to the developing shoot
tips, where it serves as the major source of soluble
nitrogen for other carbon skeletal compounds (Wickre-
masinghe and Perera, 1972; Takeo, 1979; Kito et al.,
1
970; Neumann and Montag, 1982, 1983; Konishi and
Takahashi, 1969). While most amino acids are found
at trace levels, theanine represents >50% of all amino
acids in tea. Earlier reports indicated that between 1
and 2% (on an average) of the total dry weight of tea
leaves is theanine (Cartwright et al., 1954; Wickremas-
inghe, 1978; Finger et al., 1992; Millin et al., 1969;
Konishi and Takahashi, 1966). The structure of thea-
nine is similar to that of glutamine and γ-glutamyl
dipeptides previously found in plants (see Figure 1)
1
968; Konishi and Takahashi, 1969; Konishi, 1970;
Konishi et al., 1969; Konishi and Kasai, 1968a,b).
Theanine is a significant precursor for the biosynthesis
of flavanols in tea leaves (Kito et al., 1968).
(
Sasaoka and Kito, 1964). Theanine was first found in
tea leaves (Camellia sinensis and others of the Camellia
genus) in the early 1950s (Nagashima et al., 1957). The
only other reported natural source of theanine is in the
mushroom Xerocomus badius (Casimir et al., 1960).
The flavor and fragrance components of tea can be
affected by the processing and handling steps. These
steps are listed and described in Table 1 (Lichtenstein,
1
942; Norman, 1989). Note that each tea producer can
*
Author to whom correspondence should be ad-
use some, all, or a variety of these seven basic steps.
All teas are graded on their physical form, age, and/or
dressed.
S0021-8561(96)00432-3 CCC: $14.00
© 1997 American Chemical Society

354 J. Agric. Food Chem., Vol. 45, No. 2, 1997
Ekborg-Ott et al.
Ta ble 1. Descr ip tion of th e Seven P r ocessin g a n d Ha n d lin g Step s in th e P r od u ction of Gr een , Ha lf-Gr een , a n d Bla ck
Tea s
1
2
3
. With er in g
Tea shoots are spread onto racks to dry until they become withered and limp and lose 50% of their
moisture content. This process may take a few days in the open air, but only takes a few hours in
a drum drier. All three main types of tea (green, half-green, and black) go through this process.
Green tea leaves are scalded or steamed to prevent fermentation before the rolling and drying
steps.
Nowadays, the withered tea leaves are machine rolled for an hour or two. This breaks up the cellular
structure, causing the chemical components of the leaf (oils and enzymes) to be released and
mixed together. Rolling twists the leaves, which results in a delay in the rate at which the leaves
release their essence in hot water during tea making. It is during this process the flavor,
astringency, and color of the finished tea begin to develop.
A more accurate name for this process is oxidation. For black tea processing, the green sticky tea leaves
are left for 1-4 h in a humid, cool atmosphere. This makes the tea leaves develop a coppery brown
hue, together with characteristic flavor and astringency. The fermentation step progresses as far
as each individual tea producer desires. The half-green tea process is a compromise between
black and green tea. It is fermented briefly, once before and once after rolling. The color of the tea
leaves is half-brown (half-green).
. Rollin g
. F er m en ta tion
4
5
. Dr yin g
During this process, the tea leaves are dried in temperature-controlled chambers until they turn black
and have a moisture content of ∼5%.
. Sor tin g a n d
Gr a d in g
Green teas are mainly graded on the age of the leaves and their preparation. Black and Oolong teas are
passed into vibrating machines, which sort the leaves through different mesh sizes. The tea
grade is not necessarily an indication of quality. All black and Oolong teas are “broken” to some
extent during the rolling and/or cutting steps. Since small pieces of tea leaves brew much more
quickly than large ones, consistency of size is very important. There are three basic sizes of black
and Oolong tea leaves, and these in turn are divided into grades (see Table 2).
6
7
. P a ck a gin g a n d
Sh ip p in g
Teas cannot be stored for long periods of times without loss of quality. Therefore, as soon as the tea is
sorted and graded, it is packed. Sometimes, tea is shipped in paper sacks, but usually it is packed
into plywood tea chests edged with metal for extra strength. These tea chests are mostly lined
with paper and foil to keep the tea fresh, and packs either 30 or 45 kg tea each. To keep its
aroma, loose weight tea must be stored at room temperature in an airtight container.
Once the tea has been bought at an auction, manufacturers blend their own special brand products for
the retail market. Almost all black teas on the market are made up of a mixture of different tea
types. Up to 20 or 30 different teas can be mixed into a popular blend. This is sometimes
necessary to be able to guarantee the consistency (taste, smell, quality, and price) of each tea
brand. A tea brand may vary in both taste and quality, even though it was produced at the same
tea plantation. This variation can be attributed to the climate, handling of the tea leaves,
transportation, etc. A skilled tea tester ascertains which teas are required to make up a blend
consistency, and the new tea mixture is compared to blends previously produced by the
manufacturer. Most tea blends contain three basic tea types: (1) Ceylon for flavor, (2) North
Indian for strength, and (3) African for color and brightness. Some other tea types may be added
to complete the final tea taste. In some cases, some flowers or aromatic oils have been added to
the teas for look, smell, and taste.
. Blen d in g
Ta ble 2. Gr a d es of Gr een , Ha lf-Gr een , a n d Bla ck Tea s
1
. GREEN TEAa
A. Gu n powd er Tea . This tea consists of small tight balls of young and medium-aged leaves.
B. Sen ch a Tea . This tea consists of 0.5 in. long, straight (grasslike) leaves of various ages.
C. Im per ia l Tea . This is another version of gunpowder tea, consisting of older, larger, and looser leaves.
D. Hyson Tea . This tea consists of long, twisted leaves of various ages.
. OOLONG AND BLACK TEA
2
A. Lea f Tea . This type of tea takes the longest to release its flavor. It is divided into four grades.
a . F low er y Or a n ge P ek oe is considered the finest tea available, especially if it contains “Golden tips” or “Silver tips” (new shoot tips).
The word “orange” refers to the color of the leaf tips, which are included in this grade. The word “pekoe” is Chinese for “leaf” and refers
to small tea leaves which give strong brews.
b. Or a n ge P ek oe, which has long, thin, closely twisted leaves.
c. P ek oe, which has more open leaves.
d . P ek oe Sou ch on g, which has coarse, large leaves.
B. Br oken Tea . This type of tea is ideal for quick brewing, since it tends to release its flavor more quickly than leaf tea. The three grades
available are as follows:
a . Br ok en Or a n ge P ek oe
b. Br ok en P ek oe
c. Br ok en P ek oe Sou ch on g
C. Sm a ller Lea f Tea . This type of tea is excellent to use for packaged tea and tea bags, since it brews quickly. There are three grades
available.
a . Or a n ge F a n n in gs
b. F a n n in gs consists of small parts of the broken tea leaves.
c. Du st, which is the finest size of tea particles (made from broken leaves) used in tea bags.
a
The table shows some examples of green teas. Most green teas are exported all over the world, while Sencha tea is used mainly in
J apan.
appearance. The grading systems for green, half-green
flavor compound is thought to be somewhat less, since
there is a breakdown of theanine into glutamic acid and
ethylamine after the withering and fermentation pro-
cessing steps (Feldheim et al., 1986). In other words,
the highest quality black tea is thought to contain the
lowest amount of theanine. All other amino acids
increase significantly after the withering step, since they
are breakdown products from hydrolysis of the tea
(Oolong), and black teas are given in Table 2 (Lichten-
stein, 1942; Norman, 1989). During black tea manu-
facture, the amino acids may be oxidized by catechin
o-quinones and undergo subsequent Strecker degrada-
tion, which leads to new aroma components (Finger et
al., 1992; Saijo and Takeo, 1970; Co and Sanderson,
1
970). In black tea, the importance of theanine as a

Theanine Enantiomers in Tea
J. Agric. Food Chem., Vol. 45, No. 2, 1997 355
proteins (Choudhury et al., 1980; Nagashima et al.,
the amino acid theanine in a variety of teas. To increase
the HPLC detection sensitivity for low levels of amino
acids and to simplify their isolation and identification,
fluorescent and/or UV-absorbing “tagging agents” are
often useful (Pawlowska and Armstrong, 1994; Rundlett
and Armstrong, 1994; Ekborg-Ott and Armstrong, 1996;
Carpino et al., 1986, 1972; Zukowski et al., 1992, 1993;
Pawlowska et al., 1993; Sp o¨ ndlin and K u¨ sters, 1994;
K o¨ nig, 1984; Cohen and Michaud, 1993; Cohen et al.,
1
957). There is no single identified tea aroma and flavor
constituent characteristic of green, half-green, or black
tea. Instead, the aroma and flavor of tea result from a
variety of volatile components, as well as amino acids.
In this study, the green and half-green teas were
comprised of only one tea type (i.e., tea leaves from only
one plantation and one plucking), while the black teas
were comprised of between one and three types of tea.
Amino acids are known to produce complex sensations
in humans, although the sensations of sweet, sour, salty,
bitter, and umami predominate. Of the thousands of
publications on amino acids, only a fraction deal with
their enantiomeric composition. Most R-amino acids
contain a stereogenic center and exist as both D- and
L-enantiomers. Some D-amino acids are known to have
desirable and characteristic tastes. For example, some
D-amino acids exhibit strong sweet tastes, while the
corresponding L-enantiomers (most hydrophobic L-ami-
no acids) elicit bitter tastes (Solms, 1969; Wieser et al.,
1
993; Strydom and Cohen, 1992; Schuster, 1988; Ein-
arsson et al., 1986, 1987; Carpino, 1987; Cunico et al.,
1
1
986; N a¨ sholm et al., 1987; Malmer and Schroeder,
990). In this study 9-fluorenylmethoxycarbonylglycine
chloride (FMOC-Gly-Cl) was used as an effective pre-
column derivatizing agent for theanine. Efficient enan-
tioselective separations can be done on this compound
with a γ-cyclodextrin-bonded stationary phase LC col-
umn (see Materials and Methods) (Pawlowska and
Armstrong, 1994; Rundlett and Armstrong, 1994; Ek-
borg-Ott and Armstrong, 1996; Armstrong et al.,
1
993a,b). To our knowledge, no stereoselective analyses
1
977; Kato et al., 1989). D-Tryptophan, D-phenylala-
have ever been done on theanine in tea.
nine, D-histidine, D-tyrosine, and D-leucine all exhibit
sweet tastes that are 35, 7, 7, 6, and 4 times as strong
as sucrose, respectively (Solms, 1969). Glycine and
L-alanine also exhibit strong sweet tastes. The L-
glutamic acid and L-aspartic acid are sour tasting in
their dissociated states, while their sodium salts exhibit
an umami taste. Free L-theanine and glutamic acid
have been found to be the most important umami
substances of green tea (Nagashima et al., 1957; Sakato,
MATERIALS AND METHODS
Ap p a r a tu s. The HPLC system consisted of the following
Shimadzu (Kyoto, J apan) devices: two pumps (LC-6A), a UV
detector (SPD-6A), a fluorescence detector (RF-535), and two
CR601 Chromatopac recorders. A switching valve (Rheodyne,
Cotati, CA) and a 20 µL injection valve (Rheodyne) were also
used. The system described above is also known as a “coupled
HPLC” system or “coupled column switching” system (Arm-
strong et al., 1991). The mass spectral analysis was carried
out with a FAB-MS system, Model 70 SEQ, Fisons Instru-
1
950; Sakato et al., 1950). The D- and L-enantiomers,
as well as the racemic mixture of theanine, all taste
sweet and do not seem to leave any bitter aftertastes
1
ments (Manchester, U.K.). The HNMR was performed with
(see Materials and Methods).
a J EOL FX-100 NMR (100 MHz) system (Peabody, MA).
One source of D-amino acids in mammals seems to
Colu m n s. The two columns used in this study were a C18
and a γ-cyclodextrin (γ-CD, Cyclobond II-2000), both 250 ×
be the diet (Armstrong et al., 1993a), although the action
of certain gut (intestinal) bacteria may account for their
presence in some cases (Nagata and Akino, 1990; Konno
et al., 1990). Racemization of amino acids may take
place while the foods and beverages are being processed,
stored, or prepared for human consumption (Tovar and
Schwass, 1983). Food and beverage products in which
D-amino acids seem to be prevalent are those that (1)
have been exposed to microbial activity (aging, fermen-
tation, etc.) (Ekborg-Ott and Armstrong, 1996; Br u¨ ckner
and Hausch, 1989, 1990; Gobbetti et al., 1994; Fried-
man, 1991; Liardon and Hurrell, 1983; Bunjapamai et
al., 1982; Friedman et al., 1981; Hayashi and Kameda,
4
.6 mm (i.d.) containing 5 µm support, supplied from Advanced
Separation Technologies, Inc. (Whippany, NJ ). The C18 column
was used in the reversed-phase mode, while the γ-CD column
was used in the polar-organic (also called “magic mobile
phase”) mode (Armstrong et al., 1993a,b; Pawlowska et al.,
1
993; Zukowski et al., 1993; Rundlett and Armstrong, 1994;
Ekborg-Ott and Armstrong, 1996).
Sta n d a r d s, Der iva tizin g Rea gen t, Solven ts, a n d Oth er
Accessor ies. Standard theanine samples (pure D, pure L, and
the racemic mixture) were synthesized in our laboratory from
pyroglutamic acid and ethylamine according to the method
introduced by Lichtenstein (1942) (see Figure 1). Five grams
of pyroglutamic acid (purity >99%) was added to a 100 mL
round-bottom glass flask, together with 30 mL of ethylamine.
A magnetic stir bar was added, and the flask was tightly
stoppered. The flask was placed on a magnetic stir plate, and
the mixture was stirred for 20 days at room temperature (22
1
980; Gandolfi et al., 1992; Palla et al., 1989), (2) have
been highly processed (exposed to extremes of pH, heat,
etc.) (Friedman, 1991; Liardon and Hurrell, 1983; Bun-
japamai et al., 1982; Friedman et al., 1981; Hayashi and
Kameda, 1980), and (3) contain natural sources of
D-amino acids (certain seafoods) (Preston, 1987; Corri-
gan, 1969; D’Aniello and Giuditta, 1978; Felbeck, 1985;
Matsushima et al., 1984). Several reports have been
published on the presence of D-amino acids in foods and
beverages (Ekborg-Ott and Armstrong, 1996; Pawlows-
ka and Armstrong, 1994; Rundlett and Armstrong,
(
2 °C). After 20 days, all excess ethylamine was removed
using a rotary evaporator). The white crystals (theanine)
(
obtained were recrystallized from methanol (MeOH), and the
round-bottom flask was stoppered and refrigerated overnight.
The following day, the white theanine crystals were scraped
down from the side of the flask, filtered, dried, and finally
collected into a 7 mL vial with an airtight screw cap. The
newly synthesized theanine was stored in a desiccator at room
temperature.
1
1
994; Br u¨ ckner and Hausch, 1989, 1990; Gobbetti et al.,
994; Friedman, 1991; Liardon and Hurrell, 1983;
To confirm the composition and purity, we performed
elemental analysis, mass spectrometry (see Figure 2), and
Bunjapamai et al., 1982; Friedman et al., 1981; Hayashi
and Kameda, 1980; Gandolfi et al., 1992; Palla et al.,
1
proton NMR ( H NMR) of the newly synthesized theanine. The
elemental analysis was performed by Galbraith Laboratories,
1
989).
Inc. (Knoxville, TN). Calculated from C
H, 8.10; N, 16.08; O, 27.55%. Found: C, 48.48; H, 8.11; N,
6.08; O, 27.33%. MS [FAB, matrix: glycerol + 0.1% am-
monium chloride (NH Cl)], m/ z (relative intensity) 175 [(M +
7 14 2 3
H N O : C, 48.27;
The enantiomeric purity of amino acids can be easily
and routinely determined by high-performance liquid
chromatography (HPLC). In this study we examined
the concentrations and the enantiomeric composition of
1
4
+
+
1
H) , base ion]; 176 [(M + H + 1) , 10%]. H NMR (CDCl ) δ
3

356 J. Agric. Food Chem., Vol. 45, No. 2, 1997
Ekborg-Ott et al.
F igu r e 2. Mass spectrum of synthesized theanine (molecular mass 174 g/mol).
1
(
(
.11 (t, 3H, NHCH
m, 2H, CHCH CH
t, 1H, CHCH CH
2
CH
COO), 3.20 (quartet, 2H, NHCH
CH
3
), 2.15 (m, 2H, CHCH
2
CH
2
COO), 2.4
CH ), 3.76
2
2
2
3
2
2
2
COO). Chiral LC was used to confirm
the enantiomeric purity of theanine.
We tasted the newly synthesized D,L-theanine, D-theanine,
and L-theanine to determine their palatability. The racemic
mixture, as well as the pure enantiomers, all exhibited similar,
sweet tastes, but not as sweet as sucrose. No undesirable
aftertastes were detected. This leaves the question open for
further use of theanine as a sweetening or flavoring agent.
However, currently there is no existing economical method for
producing theanine commercially on a large-scale basis. This
limits the number of studies performed on the effect of
theanine on animals and humans. In one study, theanine has
been shown to inhibit the convulsive action of caffeine in mice,
but has proven ineffective against convulsants such as strych-
nine, pentetrazole, and picrotoxin (Kimura and Murata,
1
971a,b; Kimura et al., 1975; Kimura and Murata, 1980).
FMOC-Gly-Cl was prepared in our laboratory according to
the method introduced by Carpino et al. (1986). Solvents and
organic modifiers (acetonitrile, water, acetic acid, and triethy-
lamine) were of Omnisolve or HPLC grade and supplied by
EM Science (Gibbstown, NJ ) and Fisher Scientific (St. Louis,
MO). Solid phase extraction cartridges (C18, 0.5 g), strong
cationic exchange cartridges (SCX, 600 mg), and 0.2 µm filters
F igu r e 3. Reaction of FMOC-Gly-Cl with theanine.
Armstrong, 1994; Rundlett and Armstrong, 1994; Ekborg-Ott
and Armstrong, 1996; Zukowski et al., 1992; Einarsson et al.,
1983). The method used in this study offers a detection limit
as low as the femtomole level. In aqueous buffer at pH 8.0,
no detectable racemization of the amino acid occurs during
the derivatization process (Zukowski et al., 1992). This
reagent was prepared in the laboratory according to the
directions given by Carpino et al. (1986). Derivatization was
performed by first adjusting the sample to pH 8.0 using 1%
(all disposable) were obtained from Alltech (Deerfield, IL).
Sam ple P r epar ation , Der ivatization , Ch r om atogr aph ic
Con d ition s, a n d An a lyses. All tea samples were donated
by AB Kobbs S o¨ ner (Gothenburg, Sweden). Ten grams of each
tea sample was analyzed in this study. Theanine is very
soluble in water, but not in too many other solvents at room
temperature (Lehmann and Neumann, 1974). Unfortunately,
a lot of other undesirable components of the tea samples are
also soluble in water. All tea samples required cleanup,
consisting of various extractions, before LC injections could
be done. Before derivatization with FMOC-Gly-Cl, the aque-
ous tea samples were adjusted to pH 4.0 with 6 M hydrochloric
acid (HCl) and adsorbed onto disposable SCX cartridges. The
sugars and other undesirable components (which otherwise
may interfere with the derivatization procedure by reacting
with FMOC-Gly-Cl) were washed out using 3 × 3-mL aliquots
of water. The remaining amino acids were eluted from the
SCX cartridge with a mixture of acetonitrile/sodium carbonate,
(w/v) sodium carbonate (Na
2 3
CO ). This was followed by the
addition of 30 mg of FMOC-Gly-Cl dissolved in a small amount
of acetone. The reaction vial was shaken well and allowed to
react at room temperature. See Figure 3 for the reaction of
FMOC-Gly-Cl with theanine. After 15 min, the sample was
acidified to pH 4.0 using 50% acetic acid (v/v). This stabilizes
the FMOC-Gly group. After derivatization, the sample mix-
ture was passed through a C18 solid phase extraction (SPE)
cartridge, washed with 3 × 3-mL aliquots of water, and finally
eluted with pure ethyl ether. The effluent was evaporated to
dryness and redissolved in the mobile phase to be used with
the reversed-phase HPLC system. In this study the C18/γ-
cyclodextrin (C18/γ-CD) column switching method was used to
analyze standard theanine and tea samples derivatized with
FMOC-Gly-Cl. The eluent for the C18 column was acetonitrile/
water/acetic acid/triethylamine, 300:700:2.0/0.5 (v/v/v/v), and
the flow rate was 1.0 mL/min (see Table 3 for additional
4
0:60 (v/v). Standards (pure D, pure L, and the racemic
mixture) and pretreated tea samples were then precolumn
derivatized with FMOC-Gly-Cl, as has been described in
previous papers (Armstrong et al., 1993a; Pawlowska and

Theanine Enantiomers in Tea
J. Agric. Food Chem., Vol. 45, No. 2, 1997 357
Ta ble 3. Ch r om a togr a p h ic Con d ition s Used for th e Deter m in a tion of En a n tiom er ic Ra tios of Am in o Acid s in Tea
column I (RP, C18)
column II
k′L
solute
mobile phasea
k′
column
mobile phaseb
1000/35/4/0.5
k′D
R
Rs
FMOC-Gly-theanine
300/700/2/0.5
300/700/2/0.5
6.73
5.63
γ-CD
10.49
11.96
1.14
1.68
FMOC-Gly-2-aminoadipic acid
a
Eluent: acetonitrile/water/acetic acid/triethylamine (v/v/v/v). b Eluent: acetonitrile/methanol/triethylamine/acetic acid (v/v/v/v). The
chromatographic conditions are given after column switching.
F igu r e 4. Representative chromatograms of theanine in different tea samples after derivatization with FMOC-Gly-Cl reagent.
Column: C18 (used in the reversed-phase mode; see Table 3 for chromatographic conditions). Flow rate: 1 mL/min. Detection:
UV at 266 nm. Samples (see Table 4): (A) Gunpowder (green) tea, (B) J asmine FOP (half-green) tea, and (C) Keemun FOP
(black) tea. The internal standard (FMOC-Gly-derivatized 2-aminoadipic acid) is also shown.
chromatographic conditions). Before 20 µL of the sample was
injected onto the C18 column, the tea sample was filtered
through a 0.2 µm disposable filter. A UV detector, set to
wavelength 266 nm, was used to monitor the effluent. A total
of six analyses from two different batches were performed on
each tea type [each tea batch was analyzed three times (i.e., 2
vs D-peaks). In this case, the larger peaks must often be
quantitatively measured a second time after accurate serial
dilutions. For the tea samples containing only trace amounts
of D-theanine, proper dilutions (∼5-10-fold) were used to
measure the correct amount of the D- and L-enantiomers,
respectively.
×
3 ) 6 determinations)]. Figure 4 shows typical achiral C18
Tem p er a tu r e (Ra cem iza tion ) Stu d y. To see if there
were any racemization of theanine taking place, six tea
samples (two green, two half-green, and two black teas), as
well as standard L-theanine, were heated in a water bath at
reversed-phased chromatograms of FMOC-Gly-derivatized D,L-
theanine in three different tea types [(A) Gunpowder, (B)
J asmine Flowery Orange Pekoe (FOP), and (C) Keemun FOP]
before column switching. The internal standard used, 2-ami-
noadipic acid (R-aminoadipic acid), is also shown (see Table 3
for chromatographic details). The column switching valve was
turned for 1-10 s at the maximum of the eluted peak of
interest, and a small portion of the peak (FMOC-Gly-theanine)
was introduced onto the chiral γ-CD column. This column-
switching method has been described in previous papers
4
5 ( 2 °C for 120 h. All solutions had a measured pH of 5.0.
The samples were precolumn derivatized with FMOC-Gly-Cl
and analyzed on the coupled column system (C18/γ-CD) ac-
cording to the experimental scheme mentioned above. The
increase in D-theanine was reported for each sample analyzed.
See Figure 6, Table 3, and Results and Discussion for further
information.
(Armstrong et al., 1993a; Pawlowska and Armstrong, 1994;
Hyd r olysis Stu d y. Theanine can be broken down into
glutamic acid and ethylamine. To see if pH has any significant
impact on the loss of absolute amount of theanine in nature,
hydrolysis studies were done at different pH values. Three
different tea types (Gunpowder, J asmine FOP, and Ceylon
Pekoe), as well as L- and D,L-theanine were divided into three
parts each, and adjusted to different pH values (pH 3, 7, and
11) using 0.1% triethylammonium acetate buffer. Amount of
glutamic acid formed with respect to loss in theanine [i.e.,
glutamic acid/(glutamic acid + theanine) × 100%] was reported
over a period of 336 h. Both primary amino acids (glutamic
acid and theanine) were precolumn derivatized with FMOC-
Gly-Cl. No cleanup was necessary before the samples were
analyzed on the reversed-phase C18 column using the scheme
given above. See Figure 7 and Results and Discussion for
further information.
Rundlett, and Armstrong, 1994; Ekborg-Ott and Armstrong,
1
996). The γ-CD column was used in the polar-organic mode.
The mobile phase consisted of acetonitrile/methanol/triethyl-
amine/acetic acid, 1000:35:4/0.5 (v/v/v/v), and the flow rate was
1
.0 mL/min (see Table 3). The effluent was monitored with a
fluorescence detector operated at λex ) 266 nm and λem ) 315
nm. Figure 5 shows the separation of D,L-theanine in six
different tea samples [(A) Sencha , (B) Gunpowder, (C)
Formosa Oolong, (D) Lapsang Souchong, (E) Rosen, and (F)
Keemun FOP] on the γ-CD column after column switching.
When enantiomeric ratios are determined or extremely low
levels of D-amino acids (<1%) are quantitated in the presence
of the larger amounts of the L-enantiomers, it is often inac-
curate to simply use the data from the automated peak
integrator. This is because the linear dynamic range of the
detector is sometimes exceeded for the larger peaks (i.e., L-

358 J. Agric. Food Chem., Vol. 45, No. 2, 1997
Ekborg-Ott et al.
F igu r e 5. Determination of enantiomeric composition of theanine in tea samples using FMOC-Gly-Cl derivatizing agent and
the column switching technique. Column: γ-CD (used in the polar-organic mode; see Table 3 for chromatographic conditions).
Flow rate: 1 mL/min. Detection: fluorescence, λex ) 266 nm and λem ) 315 nm. Samples (see Table 4): (A) Sencha (green) tea,
(
B) Gunpowder (green) tea, (C) Formosa Oolong (half-green) tea, (D) Lapsang Souchong (black) tea, (E) Rosen (black) tea, and (F)
Keemun FOP (black) tea.
RESULTS AND DISCUSSION
absolute amount and the enantiomeric composition of
theanine were different for each of the tea samples
examined in this study. The average amount of total
theanine was 1.37 g/100 g of tea, with Formosa Oolong
Tea is a natural product. The chemical composition
of tea can vary depending on season, climate, altitude,
tea variety, age of the leaf, position of the leaf on the
harvested shoot, horticulture practices, processing of the
plant material, shipping, handling, storage, as well as
many other factors (Graham, 1992; Konishi and Taka-
hashi, 1969). Moisture and heat will cause stored tea
to deteriorate. Loss of astringency and flavor may occur
due to lipid hydrolysis and autoxidative reactions that
cause losses of amino acids (particularly theanine),
sugars, and other flavor compounds (Stagg, 1974). To
prevent undesirable changes (loss of desirable flavors
and tastes, or addition of undesirable flavors and tastes)
tea should be stored in airtight containers.
(
0.60 g/100 g of tea) having the lowest concentration and
Yunnan (2.38 g/100 g of tea) the highest (see Table 4).
These values are in the general range of theanine levels
reported previously. Interestingly, the black teas had
roughly equivalent or higher levels of theanine com-
pared to the green and half-green teas. Thus, it is likely
that theanine imparts important flavor characteristics
to black teas, as well as to green and half-green teas.
This is contrary to some previously reported results
(Nakagawa, 1970, 1975; Nagashima et al., 1957). It is
also interesting that different grades of the same type
of tea from the same location (i.e., Ceylon Broken and
Ceylon Pekoe from Sri Lanka) can have significantly
different amounts of theanine, as well as different
enantiomeric ratios (Table 4).
Theanine has been known to play an important
enhancing role in the quality and characteristics of
green tea (Nakagawa, 1970, 1975; Nagashima et al.,
1
957). Table 4 summarizes the results obtained for the
The average relative level of D-theanine determined
in this tea study was 1.85%, with Cherry Blend (0.21%)
containing the smallest relative amount and Formosa
total amount as well as for the enantiomeric composi-
tions of theanine in 17 different tea samples. The

Theanine Enantiomers in Tea
J. Agric. Food Chem., Vol. 45, No. 2, 1997 359
F igu r e 6. Change in the enantiomeric ratio of theanine
during heating in a water bath at 45 ( 2 °C, using the column
switching method. The chromatographic conditions used for
the determinations are listed in Table 3. (A) Samples (see
Table 4): (×) L-theanine standard, (9) Gunpowder (green) tea,
(
4) Sencha (green) tea, (2) J asmine (half-green) tea, (b) Ceylon
Pekoe (black) tea, and (O) Darjeeling FOP (black) tea. (B)
Sample: ([) Formosa Oolong (half-green) tea.
Oolong (12.7%) the largest. Formosa Oolong is an
interesting tea. In addition to having the highest
amount of the D-enantiomer of theanine, it also con-
tained the smallest absolute amount of theanine (0.60
g/100 g of tea). A possible explanation for this distribu-
tion may be found in the handling of the Formosa
Oolong tea. Formosa Oolong is a half-green tea, con-
sisting of only one type of tea (tea leaves from only one
plantation). Since the Oolong tea is partially oxidized
during the fermentation step, there may be some
restricted growth of microorganisms, such as yeasts or
bacteria in the tea leaves (Harler, 1963; Graham, 1992).
These microorganisms may either utilize the L-theanine
as a source of energy (leaving behind the accumulated
D-enantiomer) or racemize the L-theanine while it is
being used (i.e., microbial fermentation). Another pos-
sible explanation may be the degradation of L-theanine
by tea enzymes. More extensive studies will have to
be done to determine the reason for the high percentage
of D-theanine in Formosa Oolong tea.
F igu r e 7. Representative graphs of the hydrolysis experiment
performed at various pH values. The change (increase) in
absolute amount of glutamic acid is shown. Samples (see Table
4): (O) D,L-theanine standard, (×) L-theanine standard, (9)
Gunpowder (green) tea, (2) J asmine FOP (half-green) tea, and
(
b) Ceylon Pekoe (black) tea. Hydrolysis was at (A) pH 3, (B)
pH 7, and (C) pH 11.
“pekoe” is Chinese for “leaf” and refers to small tea
leaves which give strong brews. FOP is considered the
finest tea available, especially if it contains new shoot
tips (golden tips or silver tips). The word “orange” refers
to the color of the leaf tips. It may take many years to
become a proficient tea grader. The position of the
An interesting trend relating to the enantiomeric
composition of theanine was found in Table 4. The teas
that had the lowest relative amounts of D-theanine
(<1%) were always of pekoe or FOP grades. The word

360 J. Agric. Food Chem., Vol. 45, No. 2, 1997
Ekborg-Ott et al.
Ta ble 4. En a n tiom er ic Com p osition a n d Tota l Am ou n t of th e Am in o Acid Th ea n in e in Va r iou s Tea Sa m p les
country where
produced
theanine
%Da
total amountc,d
(g of theanine/100 g of tea)
tea name
tea type
SDb
e
African Flower
Assam FOPf
Ceylon Broken
Ceylon Pekoe
black
black
black
black
black
black
black
half-green
black
green
half-green
black
black
black
black
green
black
Kenya
India
Sri Lanka
Sri Lanka
India/China
India
0.54
0.49
2.30
0.34
0.21
0.45
0.42
12.7
0.46
2.20
0.45
0.65
1.04
2.70
2.46
2.20
1.79
0.05
0.03
0.09
0.04
0.04
0.03
0.03
0.4
0.02
0.20
0.04
0.06
0.12
0.23
0.24
0.09
0.16
1.30
1.05
1.32
2.20
2.04
1.45
1.07
0.60
1.16
1.78
1.72
1.12
0.82
1.26
1.03
1.05
2.38
f
Cherry Blend FOP
Darjeeling FOP
f
e
Earl Grey
China
Formosa Oolong
Georgian FOPf
Gunpowder
Taiwan
Georgia
China
China
China
China
India/Sri Lanka
China
J apan
China
f
J asmine FOP
Keemun FOPf
Lapsang Souchong
Lemon Blend
Rosen
Sencha
Yunnan
a
The number shown is the mean of six separate enantiomeric analyses performed on each tea type [i.e., D/(D + L) × 100%] using a
b
γ-CD column. The average deviation form the mean for six separate analyses of D-theanine. Three determinations from two different
c
tea batches were performed (i.e., 3 × 2 ) 6 determinations). This number represents the total amino acid content regardless of chirality
(
i.e., L plus D enantiomers) and is the average of six determinations performed on a C18 column. (Three determinations from two different
d
tea batches were performed.)
2-Aminoadipic acid (R-aminoadipic acid) was used as the internal standard for all quantitative
e
measurements. The low relative level of D-theanine seemed to indicate that African Flower and Earl Grey are either pekoe or FOP
grade teas. Indeed, further investigation revealed that they are FOP teas. FOP, Flowery Orange Pekoe.
f
Ta ble 5. Va r ia tion s in Tw o Differ en t Lots of th e Sa m e
Tea Reflect Differ en ces in Tota l a n d En a n tiom er ic
Am ou n ts of Th ea n in e
to-batch variations occur within the same type of tea.
Formosa Oolong tea showed the most variability in the
absolute amount of theanine (0.30 g/100 g of tea).
Sencha tea and Yunnan tea had differences of 0.25 and
theanine
0
.17 g/100 g of tea, respectively, in the absolute amounts
range of total
range of
D-theaninee
(%)
a-c
amount
of theanine. The enantiomeric differences were 1.2%
for Formosa Oolong tea, 0.40% for Yunnan tea, and
tea type
Sencha
Formosa Oolong
Yunnan
(g/100 g of tea) SDd
SDc,d
0
.23% for Sencha tea. Consequently, one should expect
0.93-1.18
0.43-0.73
2.29-2.46
0.09
0.11
0.06
2.09-2.32
12.1-13.3
1.58-1.98
0.09
0.4
0.16
at least this much variation in the theanine concentra-
tions (both absolute and enantiomeric) in different types
of tea. As can be seen in Table 4, the variation in
different types and grades of teas generally exceeds that
found between batches of the same type (and grade) of
tea from the same location. There can be as much as a
a
These numbers represent the range for total amount of
theanine regardless of chirality (i.e., L plus D enantiomers) for six
samples analyzed on a C18 column. (Three determinations from
two different tea batches were performed.) 2-Aminoadipic acid
b
was used as the internal standard for the quantitative measure-
4
-fold difference in the absolute amounts of theanine
c
ments. The chromatographic conditions are the same as those
d
in different tea types and as much as a 60-fold difference
in relative amounts of D-theanine.
listed in Table 3. This number is the average deviation from the
mean for six separate determinations. Each tea batch was
e
analyzed three times (i.e., 2 × 3 ) 6 determinations). These
The significant amounts of the D-enantiomer found
in a number of commercially available teas suggest that
enantiomeric ratios might be useful as an indicator for
long-term storage, or possibly for shipping, handling,
and processing procedures of each type of tea. This
could be particularly true for tea-based drinks that are
sold already brewed and bottled. To support this
hypothesis (to see if solution racemization of theanine
takes place), six tea samples (two green, two half-green,
and two black teas), as well as standard L-theanine,
were heated in a water bath at 45 ( 2 °C for a measured
period of time. The results reported in Figure 6 show
significant increase in the level of D-theanine in all
samples analyzed. Linear relationships were found (D-
theanine vs time) for all samples studied, although the
racemization of theanine was much lower in the stan-
dard L-theanine sample compared to any of the “real-
world” teas. This suggests that other tea factors are
involved in the racemization of theanine in teas.
numbers represent the range for the enantiomeric composition of
theanine (i.e. D/(D + L) × 100%) for six analyses performed on a
γ-CD column. Three determinations from two different tea batches
were performed.
leaves on the tea plant, the size of the tea leaves, and
their physical appearance must be taken into account
when teas are graded. It appears from the results
(
Table 4) that there may be a quantitative relationship
between the grade of tea and the relative level of
D-theanine. Generally, the pekoe and FOP grades of
tea contained the lowest percent of D-theanine.
We also examined the natural variation of theanine
within a single type of tea coming from the same
producer (i.e., lot-to-lot or batch-to-batch variations).
Table 5 summarizes the results for total amount and
enantiomeric composition of theanine in three different
tea types: Sencha (green) tea, Formosa Oolong (half-
green) tea, and Yunnan (black) tea. Sencha is repre-
sentative of a green tea. Formosa Oolong tea was
chosen because it tended to have the lowest absolute
amount of theanine and the highest relative amount of
D-theanine. The Yunnan tea was chosen since it had
the highest absolute amount of theanine of all samples
analyzed. Different batch samples of each tea were
analyzed three times each. The results show that batch-
In nature, theanine is also broken down into glutamic
acid and ethylamine. Consequently, the hydrolysis of
theanine was examined at different pH values. Three
different tea types (Gunpowder, J asmine FOP, and
Ceylon Pekoe), as well as L- and D,L-theanine were
divided into three parts each and adjusted to different
pH values (pH 3, 7, and 11). Analyses were done

Theanine Enantiomers in Tea
J. Agric. Food Chem., Vol. 45, No. 2, 1997 361
regularly over a period of 336 h. This study measured
the amount of glutamic acid produced with respect to
loss of theanine [i.e., glutamic acid/(glutamic acid +
theanine) × 100%] vs time. Figure 7 summarizes the
results of the hydrolysis study. There seems to be a
linear relationship for glutamic acid formation with time
at different pH values. There was little difference in
hydrolysis of theanine between pH 3 and 7. However,
the samples at pH 11 showed a far greater amount of
hydrolysis. The standard theanine samples, dissolved
in buffered aqueous solutions, did not show as much
hydrolysis as the individual tea samples. Once again,
this indicates that there may be other components in
tea that accentuate reactions involving theanine (in this
case, hydrolysis).
Br u¨ ckner, H.; Hausch, M. Gas chromatographic detection of
D-amino acids as common constituents of fermented foods.
Chromatographia 1989, 28 (9-10), 487-492.
Br u¨ ckner, H.; Hausch, M. D-Amino acids as ubiquitous con-
stituents in fermented foods. In Amino Acids: Chemistry,
Biology and Medicine; Lubec, G., Rosenthal, G. A., Eds.;
ESCOM Scientific: Leiden, The Netherlands, 1990; pp
1
172-1182.
Bunjapamai, S.; Mahoney, R. R.; Fagerson, I. S. Determination
of D-amino acids in some processed foods and effects of
racemization on in vitro digestibility of casein. J . Food Sci.
1982, 47 (4), 1229-1234.
Carpino, L. A. The 9-fluorenylmethyloxycarbonyl family of
base-sensitive amino-protecting groups. Acc. Chem. Res.
1
987, 20 (11), 401-407.
Carpino, L. A.; Han, G. Y. 9-Fluorenylmethoxycarbonyl amino-
protecting group. J . Org. Chem. 1972, 37 (22), 3404-3409.
Carpino, L. A.; Cohen, B. J .; Stephens, K. E., J r.; Sadat-Aalaee,
S. Y.; Tien, J . H.; Langridge, D. C. ((9-Fluorenylmethyl)-
oxy)carbonyl (Fmoc) amino acid chlorides. Synthesis, char-
acterization, and applications to the rapid synthesis of short
peptide segments. J . Org. Chem. 1986, 51 (19), 3732-3734.
Cartwright, R. A.; Roberts, E. A. H.; Wood, D. J . Theanine,
an amino acid N-ethyl amide present in tea. J . Sci. Food
Agric. 1954, 5, 597-599.
CONCLUSIONS
Chiral discrimination plays an important role in odor,
aroma, and taste perception, as well as in biological
structure-function relationships. Free D-theanine was
found at significant levels in all tea samples analyzed
in this study. Contrary to previous reports, theanine
was as prevalent in black and Oolong teas as in green
teas. Therefore, theanine, which is known as an
important taste element in green tea, must also play
an important role in the taste characteristics of Oolong
and black teas. Each of the pure enantiomers and the
racemic mixture of theanine have a similar, sweet taste,
with no bitter aftertaste.
There does not seem to be any correlation between
the absolute amount of theanine in tea and its enan-
tiomeric composition, with the possible exception of
Formosa Oolong. On the other hand, there appears to
be a correlation between enantiomeric composition of
theanine and certain grades of tea. Lower percentages
Casimir, J .; J adot, J .; Renard, M. Separation and characteriza-
tion of N-ethyl-γ-glutamine in Xerocomus badius (Boletus
ladius). Biochim. Biophys. Acta 1960, 39, 462-468.
Chakraborty, S.; Srivastava, R. A. K.; Choudhury, M. N. Dev.
Distribution of free amino acids and chlorogenic acids in
different parts of “two leaves and a bud” and its relation to
tea quality. Two Bud 1978, 25 (1), 17-21.
Choudhury, M. N. Dev; Bajaj, K. L. Role of chlorophylls, amino
acids and sugars in tea. Two Bud 1980, 27 (1), 16-20.
Co, H.; Sanderson, G. W. Biochemistry of tea fermentation:
conversion of amino acids to black tea aroma constituents.
J . Food Sci. 1970, 35 (2), 160-164.
Cohen, S. A.; Michaud, D. P. Synthesis of a fluorescent
derivatizing reagent, 6-aminoquinolyl-N-hydroxysuccinim-
idyl carbamate, and its application for the analysis of
hydrolysate amino acids via high-performance liquid chro-
matography. Anal. Biochem. 1993, 211 (2), 279-287.
Cohen, S. A.; DeAntonis, K.; Michaud, D. P. Compositional
protein analysis using 6-aminoquinolyl-N-hydroxysuccin-
imidyl carbamate, a novel derivatization reagent. In Tech-
niques in Protein Chemistry 4; Angeletti, R. H., Ed.;
Academic: San Diego, CA, 1993; pp 289-298.
(<1%) of the D-enantiomer of theanine were always
associated with pekoe or FOP teas. This suggests that
enantiomeric analysis may be a useful tool in the
grading of teas.
The hydrolysis of theanine is accelerated in basic
solutions. Theanine also racemizes in aqueous solution.
The theanine in authentic tea samples both racemized
and hydrolyzed more quickly than did standards of
theanine dissolved in aqueous or buffered solutions.
This could be useful in evaluating storage, handling, and
shipping practices for tea and tea-based products.
Corrigan, J . J . D-Amino acids in animals. Science 1969, 164
(3876), 142-149.
Crocomo, O. J .; Fowden, L. Amino acid decarboxylases of
higher plants: formation of ethylamine. Phytochemistry
1
970, 9 (3), 537-540.
Cunico, R.; Mayer, A. G.; Wehr, C. T.; Sheehan, T. L. High
sensitivity amino acid analysis using a novel automated
precolumn derivatization system. BioChromatography 1986,
ACKNOWLEDGMENT
1
(1), 6-14.
Tea samples (loose weight) of various geographic
origins were generously donated from AB Kobbs S o¨ ner
D’Aniello, A.; Giuditta, A. Presence of D-aspartate in squid
axoplasm and in other regions of the cephalopod nervous
system. J . Neurochem. 1978, 31 (4), 1107-1108.
Einarsson, S.; J osefsson, B.; Lagerkvist, S. Determination of
amino acids with 9-fluorenylmethyl chloroformate and
revered-phase high-performance liquid chromatography. J .
Chromatogr. 1983, 282, 609-618.
(
Gothenburg, Sweden). We thank Professors Shub-
hender Kapila and Frank Blum, who helped with the
MS and the HNMR analyses, respectively, that were
1
used in the identification of theanine.
Einarsson, S.; Folestad, S.; J osefsson, B.; Lagerquist, S. High-
resolution reversed-phase liquid chromatography system for
the analysis of complex solutions of primary and secondary
amino acids. Anal. Chem. 1986, 58 (8), 1638-1643.
Einarsson, S.; J osefsson, B.; M o¨ ller, P.; Sanchez, D. Separation
of amino acid enantiomers and chiral amines using precol-
umn derivatization with (+)-1-(9-fluorenyl)ethyl chlorofor-
mate and reversed-phase liquid chromatography. Anal.
Chem. 1987, 59 (8), 1191-1195.
Ekborg-Ott, K. H.; Armstrong, D. W. Evaluation of the
concentration and enantiomeric purity of selected free amino
acids in fermented malt beverages (beers). Chirality 1996,
8 (1), 49-57.
LITERATURE CITED
Armstrong, D. W.; Duncan, J . D.; Lee, S. H. Evaluation of
D-amino acid levels in human urine and in commercial
L-amino acid samples. Amino Acids 1991, 1 (1), 97-106.
Armstrong, D. W.; Gasper, M. P.; Lee, S. H.; Ercal, N.;
Zukowski, J . Factors controlling the level and determination
of D-amino acids in the urine and plasma of laboratory
rodents. Amino Acids 1993a , 5 (2), 299-315.
Armstrong, D. W.; Gasper, M.; Lee, S. H.; Zukowski, J .; Ercal,
N. D-Amino acids in human physiological fluids. Chirality
1
993b, 5 (5), 375-378.

362 J. Agric. Food Chem., Vol. 45, No. 2, 1997
Ekborg-Ott et al.
1
4
Felbeck, H. Occurrence and metabolism of D-aspartate in the
gutless bivalve Solemya reidi. J . Exp. Zool. 1985, 234 (1),
metabolic changes of carbon- C dioxide assimilated prod-
ucts in tea plants during autumn. Nippon Dojo-Hiryogaku
Zasshi 1968b, 39 (9), 444-446.
1
45-149.
Feldheim, W.; Yongvanit, P.; Cummings, P. H. Investigation
of the presence and significance of theanine in the tea plant.
J . Sci. Food Agric. 1986, 37 (6), 527-534.
Finger, A.; Kuhr, S.; Engelhardt, U. H. Chromatography of
tea constituents. J . Chromatogr. 1992, 624 (1-2), 293-315.
Friedman, M. Formation, nutritional value and safety of
D-amino acids. In Nutritional and Toxicological Conse-
quences of Food Processing; Friedman, M., Ed.; Plenum:
New York, 1991; pp 447-481.
Konishi, S.; Takahashi, E. Existence and synthesis of L-
glutamic acid γ-methylamide in tea plants. Plant Cell
Physiol. 1966, 7 (1), 171-175.
Konishi, S.; Takahashi, E. Metabolism of theanine-N-ethyl-
1
4
C and its metabolic redistribution in the tea plant. VI.
Metabolism and regulation of theanine and related com-
pounds in the tea plant. Nippon Dojo-Hiryogaku Zasshi
1969, 40 (11), 479-484.
Konishi, S.; Matsuda, T.; Takahashi, E. Synthesis of theanine
and L-glutamic acid γ-methylamide in Thea sinensis, Camel-
lia sasanqua, and Oryza sativa. V. Metabolism and regula-
tion of theanine and related compounds in the tea plant.
Nippon Dojo-Hiryogaku Zasshi 1969, 40 (3), 107-112.
Konno, R.; Niwa, A.; Yasumura, Y. Intestinal bacterial origin
of D-alanine in urine of mutant mice lacking D-amino acid
oxidase. Biochem. J . 1990, 268 (2), 263-265.
Lehmann, G.; Neumann, B. Theanine content of tea. Z.
Lebensm. Unters. Forsch. 1974, 156 (1), 32-35.
Liardon, R.; Hurrell, R. F. Amino acid racemization in heated
and alkali-treated proteins. J . Agric. Food Chem. 1983, 31
Friedman, M.; Zahnley, J . C.; Masters, P. M. Relationship
between in vitro digestibility of casein and its content of
lysinoalanine and D-amino acids. J . Food Sci. 1981, 46 (1),
1
27-131, 134.
Gandolfi, I.; Palla, G.; Delprato, L.; De Nisco, F.; Marchelli,
R.; Salvadori, C. D-Amino acids in milk as related to heat
treatments and bacterial activity. J . Food Sci. 1992, 57 (2),
3
77-379.
Gobbetti, M.; Simonetti, M. S.; Rossi, J .; Cossignani, L.;
Corsetti, A.; Damiani, P. Free D- and L-amino acid evolution
during sourdough fermentation and baking. J . Food Sci.
1
994, 59 (4), 881-884.
(2), 432-437.
Graham, H. N. Green tea composition, consumption, and
polyphenol chemistry. Prev. Med. 1992, 21 (3), 334-350.
Gray, O. D. Thesis, University of London, 1963.
Harler, C. R. In Tea Manufacture; Oxford University: London,
U.K., 1963.
Lichtenstein, N. Preparation of γ-alkylamides of glutamic acid.
J . Am. Chem. Soc. 1942, 64, 1021-1022.
Malmer, M. F.; Schroeder, L. A. Amino acid analysis by high-
performance liquid chromatography with methanesulfonic
acid hydrolysis and 9-fluorenylmethylchloroformate deriva-
tization. J . Chromatogr. 1990, 514 (2), 227-239.
Matsushima, O.; Katayama, H.; Yamada, K.; Kado, Y. Occur-
rence of free D-alanine and alanine racemase activity in
bivalve molluscs with specific reference to intracellular
osmoregulation. Mar. Biol. Lett. 1984, 5 (4), 217-225.
Matsuura, T.; Kakuda, T. Effects of precursor, temperature,
and illumination on theanine accumulation in tea. Agric.
Biol. Chem. 1990, 54 (9), 2283-2286.
Millin, D. J .; Crispin, D. J .; Swaine, D. Nonvolatile components
of black tea and their contribution to the character of the
beverage. J . Agric. Food Chem. 1969, 17 (4), 717-722.
Nagashima, Z.; Nakagawa, M.; Tokumura, H.; Toriumi, Y.
Free amino acids in tea. IV. Quantitative changes in the
manufacture of black and green tea. J . Agric. Chem. Soc.
J pn. 1957, 31, 169-173.
Nagata, Y.; Akino, T. D-Amino acids in mouse tissues are not
of microbial origin. Experientia 1990, 46 (5), 466-468.
Nakagawa, M. Constituents in tea leaf and their contribution
to the taste of green tea liquor. J pn. Agric. Res. Q. 1970, 5
Hayashi, R.; Kameda, I. Racemization of amino acid residues
during alkali-treatment of protein and its adverse effect of
pepsin digestibility. Agric. Biol. Chem. 1980, 44 (4), 891-
8
95.
Kato, H.; Rhue, M. R.; Nishimura, T. Role of free amino acids
and peptides in food taste. In Flavor ChemistrysTrends and
Developments; Teranishi, R., Buttery, R. G., Shahidi, F.,
Eds.; ACS Symposium Series 388; American Chemical
Society: Washington, DC, 1989; pp 158-174.
Kawamura, Y., Kare, M. R., Eds. In Umami: A Basic Taste:
Biochemistry, Nutrition, Food Science; Dekker: New York,
1
987; p 649.
Kimura, R.; Murata, T. Influence of alkylamides of glutamic
acid and related compounds on the central nervous system.
I. Central depressant effect of theanine. Chem. Pharm. Bull.
1
971a , 19 (6), 1257-1261.
Kimura, R.; Murata, T. Influence of alkylamides of glutamic
acid and related compounds on the central nervous system.
II. Syntheses of amides of glutamic acid and related
compounds, and their effects on the central nervous system.
Chem. Pharm. Bull. 1971b, 19 (7), 1301-1307.
(3), 43-47.
Nakagawa, M. Chemical components and taste of green tea.
J pn. Agric. Res. Q. 1975, 9 (3), 156-160.
Kimura, R.; Murata, T. Influence of alkylamides of glutamic
acid and related compounds on the central nervous system.
IV. Effect of theanine on adenosine 3′,5′-monophosphate
formation in rat cerebral cortex. Chem. Pharm. Bull. 1980,
N a¨ sholm, T.; Sandberg, G.; Ericsson, A. Quantitative analysis
of amino acids in conifer tissues by high-performance liquid
chromatography and fluorescence detection of their 9-fluo-
renylmethyl chloroformate derivatives. J . Chromatogr. 1987,
2
8 (2), 664-666.
Kimura, R.; Kurita, M.; Murata, T. Influence of alkylamides
of glutamic acid and related compounds on the central
nervous system. III. Effects of theanine on the spontaneous
activity of mice. J . Pharm. Soc. J pn. 1975, 95 (7), 892-895.
Kito, M.; Kokura, H.; Izaki, J .; Sasaoka, K. Theanine, a
precursor of the phloroglucinol nucleus of catechins in the
tea plants. Phytochemistry 1968, 7 (4), 599-603.
3
96, 225-236.
Neumann, K.; Montag, A. Quantitative determination of
theanine in tea extracts. Dtsch. Lebensm.-Rundsch. 1982,
7
8 (5), 172-174.
Neumann, K.; Montag, A. Study of some nitrogenous sub-
stances in tea (Camellia sinensis). Dtsch. Lebensm.-Rund-
sch. 1983, 79 (5), 160-164.
K o¨ nig, W. A. New developments in enantiomeric separation
by capillary gas chromatography. In Analysis of Volatiless
Methods and Applications; Schreier, P., Ed.; de Gruyter:
Berlin, Germany, 1984; pp 77-91.
Konishi, S. Physiological chemistry on two amides contained
in the tea tree. Chagyo Kenkyu Hokoku Shiryo 1970, No. 2,
Norman, J . In Teas and Tisanes; Edmonds, G., Ed.; Dorling
Kindersley: London, U.K., 1989.
Orihara, Y.; Furuya, T. Production of theanine and other
γ-glutamyl derivatives by Camellia sinensis cultured cells.
Plant Cell Rep. 1990, 9 (2), 65-68.
2
2-32.
Palla, G.; Marchelli, R.; Dossena, A.; Casnati, G. Occurrence
of D-amino acids in food. Detection by capillary gas chro-
matography and by reversed-phase high-performance liquid
chromatography with L-phenylalaninamides as chiral selec-
tors. J . Chromatogr. A 1989, 475, 45-53.
Pawlowska, M.; Armstrong, D. W. Evaluation of enantiomeric
purity of selected amino acids in honey. Chirality 1994, 6
(4), 270-276.
Konishi, S.; Kasai, Z. Metabolism and regulation of theanine
and related compounds in tea plants. II. Synthesis of
theanine from carbon- C dioxide in tea plants and sites of
the synthesis. Nippon Dojo-Hiryogaku Zasshi 1968a , 39 (9),
4
1
4
39-443.
Konishi, S.; Kasai, Z. Metabolism and regulation of theanine
and related compounds in tea plants. III. Translocation and

Theanine Enantiomers in Tea
J. Agric. Food Chem., Vol. 45, No. 2, 1997 363
Pawlowska, M.; Chen, S.; Armstrong, D. W. Enantiomeric
separation of fluorescent, 6-aminoquinolyl-N-hydroxysuc-
cinimidyl carbamate, tagged amino acids. J . Chromatogr.
Stagg, G. V. Chemical changes occurring during the storage
of black tea. J . Sci. Food Agric. 1974, 25 (8), 1015-1034.
Strydom, D. J .; Cohen, S. A. Bioforum, No. 3; Millipore:
Bedford, MA, 1992.
Takeo, T. L-Alanine as a precursor of ethylamine in Camellia
sinensis. Phytochemistry 1974, 13 (8), 1401-1406.
Takeo, T. Formation of amino acids induced by ammonia
application and seasonal level fluctuation of amino acids
contents in tea plant. Study Tea 1979, No. 56, 70-77.
Teranishi, R. New trends and developments in flavor chem-
istry. In Flavor ChemistrysTrends and Developments; Ter-
anishi, R., Buttery, R. G., Shahidi, F., Eds.; ACS Symposium
Series 388; American Chemical Society: Washington, DC,
1
993, 641 (2), 257-265.
Perera, K. P. W. C.; Wickremasinghe, R. L. Role of theanine
in tea manufacture. Proc. Ceylon Assoc. Adv. Sci. 1971, 1,
8
0.
Preston, R. L. Occurrence of D-amino acids in higher organ-
isms: a survey of the distribution of D-amino acids in marine
invertebrates. Comp. Biochem. Physiol., B: Comp. Biochem.
1
987, 87B (1), 55-62.
Roberts, E. A. H. Chemistry of tea fermentation. J . Sci. Food
Agric. 1952, 3, 193-198.
Rundlett, K. L.; Armstrong, D. W. Evaluation of free D-
glutamate in processed foods. Chirality 1994, 6 (4), 277-
1
989; pp 1-6.
Tovar, L. R.; Schwass, D. E. D-Amino acids in processed
proteins: their nutritional consequences. In Xenobiotics in
Foods and Feeds; Finley, J . W., Schwass, D. E., Eds.; ACS
Symposium Series 234; American Chemical Society: Wash-
ington, DC, 1983; pp 169-185.
Tunmann, P.; Linde, H. Nitrogen-containing contents of the
roots of Bryonia dioica. Arch. Pharm. 1958, 291, 263-268.
Wickremasinghe, R. L. Tea [in food research]. Adv. Food Res.
2
82.
Saijo, R.; Takeo, T. Formation of aldehydes from amino acids
by tea leaves extracts. Agric. Biol. Chem. 1970, 34 (2), 227-
2
33.
Sakato, Y. The chemical constituents of tea. III. A new amide,
theanine. J . Agric. Chem. Soc. J pn. 1950, 23, 262-267.
Sakato, Y.; Hashizume, T.; Kishimoto, Y. The chemical con-
stituents of tea. V. Synthesis of theanine. J . Agric. Chem.
Soc. J pn. 1950, 23, 269-271.
1
978, 24, 229-286.
Wickremasinghe, R. L.; Perera, K. P. W. C. Site of biosynthesis
and translocation of theanine in the tea plant. Tea Q. 1972,
Sasaoka, K. Biosynthesis of theanine. J . Agric. Chem. Soc. J pn.
1
965, 39 (11), R1-R6.
4
3 (4), 175-179.
Sasaoka, K.; Kito, M. Synthesis of theanine by tea seedling
homogenate. Agric. Biol. Chem. 1964, 28 (5), 313-317.
Sasaoka, K.; Kito, M.; Inagaki, H. Biosynthesis of theanine in
tea seedlings. Synthesis of theanine homogenate of tea
seedlings. Agric. Biol. Chem. 1963, 27, 467-468.
Wieser, H.; J ugel, H.; Belitz, H. D. Relationships between
structure and sweet taste of amino acids. Z. Lebensm.
Unters. Forsch. 1977, 164 (4), 277-282.
Zukowski, J .; Pawlowska, M.; Armstrong, D. W. Efficient
enantioselective separation and determination of trace
impurities in secondary amino acids (i.e., imino acids). J .
Chromatogr. 1992, 623 (1), 33-41.
Zukowski, J .; Pawlowska, M.; Nagatkina, M.; Armstrong, D.
W. High-performance liquid chromatographic enantiosepa-
ration of glycyl di- and tripeptides on native cyclodextrin
bonded phases. J . Chromatogr. 1993, 629 (2), 169-179.
Sasaoka, K.; Kito, M.; Onishi, Y. Some properties of the
theanine synthesizing enzyme in tea seedlings. Agric. Biol.
Chem. 1965, 29 (11), 984-988.
Schuster, R. Determination of amino acids in biological,
pharmaceutical, plant and food samples by automated
precolumn derivatization and high-performance liquid chro-
matography. J . Chromatogr. 1988, 431 (2), 271-284.
Selvendran, R. R. Changes in the composition of the xylem
exudate of tea plants (Camellia sinensis) during recovery
from pruning. Ann. Bot. 1970, 34 (137), 825-833.
Serenkov, G. P. Ammonium and volatile amine of tea seeds.
Nauchn. Dokl. Vyssh. Shk., Biol. Nauki 1959, 2, 175-179.
Solms, J . Taste of amino acids, peptides, and proteins. J . Agric.
Food Chem. 1969, 17 (4), 686-688.
Received for review J une 17, 1996. Revised manuscript
X
received Novmber 18, 1996. Accepted November 22, 1996.
Support of this work by the Environmental Protection Agency
(R823360-01-0) is gratefully acknowledged.
J F960432M
Sp o¨ ndlin, Ch.; K u¨ sters, E. Direct determination of enantio-
meric purity of FMOC amino acids with high performance
liquid chromatography. Chromatographia 1994, 38 (5-6),
X
Abstract published in Advance ACS Abstracts, Janu-
3
25-329.
ary 15, 1997.