Chemoenzymatic synthesis of aroma active 5,6-dihydro- and tetrahydropyrazines from aliphatic acyloins produced by baker's yeast

Article abstract of DOI:10.1021/jf0261809

Twenty-five acyloins were generated by biotransformation of aliphatic aldehydes and 2-ketocarboxylic acids using whole cells of baker's yeast as catalyst. Six of these acyloins were synthesized and tentatively characterized for the first time. Subsequent chemical reaction with 1,2-propanediamine under mild conditions resulted in the formation of thirteen 5,6-dihydropyrazines and six tetrahydropyrazines. Their odor qualities were evaluated, and their odor thresholds were estimated. Among these pyrazine derivatives, 2-ethyl-3,5-dimethyl-5,6-dihydropyrazine (roasted, nutty, 0.002 ng/L air), 2,3-diethyl-5-methyl-5,6-dihydropyrazine (roasted, 0.004 ng/L air), and 2-ethyl-3,5-dimethyltetrahy-dropyrazine (bread crustlike, 1.9 ng/L air) were the most intensive-smelling aroma active compounds.

Full text of DOI:10.1021/jf0261809

J. Agric. Food Chem. 2003, 51, 3103−3107 3103
Chemoenzymatic Synthesis of Aroma Active 5,6-Dihydro- and
Tetrahydropyrazines from Aliphatic Acyloins Produced by
Baker’s Yeast
TOSHINARI KURNIADI,† RACHID BEL RHLID,† LAURENT-BERNARD FAY,†
‡
MARCEL-ALEXANDRE JUILLERAT,*,† AND RALF GU¨ NTER BERGER
Nestle´ Research Center, Vers-chez-les-Blanc, P.O. Box 44, 1000 Lausanne 26, Switzerland, and
Universita¨t Hannover, Wunstorfer Strasse 14, 30453 Hannover, Germany
Twenty-five acyloins were generated by biotransformation of aliphatic aldehydes and 2-ketocarboxylic
acids using whole cells of baker’s yeast as catalyst. Six of these acyloins were synthesized and
tentatively characterized for the first time. Subsequent chemical reaction with 1,2-propanediamine
under mild conditions resulted in the formation of thirteen 5,6-dihydropyrazines and six tetrahydro-
pyrazines. Their odor qualities were evaluated, and their odor thresholds were estimated. Among
these pyrazine derivatives, 2-ethyl-3,5-dimethyl-5,6-dihydropyrazine (roasted, nutty, 0.002 ng/L air),
2,3-diethyl-5-methyl-5,6-dihydropyrazine (roasted, 0.004 ng/L air), and 2-ethyl-3,5-dimethyltetrahy-
dropyrazine (bread crustlike, 1.9 ng/L air) were the most intensive-smelling aroma active compounds.
KEYWORDS: Pyrazine derivatives; acyloins; baker’s yeast; carboligation; flavor
INTRODUCTION
Alkylpyrazines are aroma active molecules that impart nutty,
roasted, or earthy tonalities (1-3). Their identification, char-
acterization, synthesis, and application in food products have
been widely studied (4-6). Alkylpyrazines have been identified
in plants (7), insects (8), and fermented and processed foods
(6). In fermented foodstuffs, such as soy sauce, sake, and vine-
gar, pyrazines arise from microbial processes (9). In processed
foods, the generation of alkylpyrazines is well-known to be
associated with the Maillard reaction (10) or pyrolysis reactions
(e.g., roasted coffee) (11). The closely related 5,6-dihydropy-
razines, however, have received little attention. They have been
produced from diketones and diamines at ambient temperatures
(12) and described as precursors of the corresponding pyrazines
(13, 14). A systematic study of the aroma qualities and aroma
thresholds of these compounds has not been undertaken.
Tetrahydropyrazines have never been reported as aroma active
compounds so far.
In the present work, thirteen 5,6-dihydropyrazines and six
tetrahydropyrazines were generated by a two step chemoenzy-
matic approach. In the first step, a pool of acyloins was obtained
by biotransformation of aliphatic aldehydes and 2-ketocarboxylic
acids using whole cells of baker’s yeast as the catalyst. The
second step consisted of a chemical reaction of these acyloins
with 1,2-propanediamine under mild conditions. Some of the
generated pyrazine derivatives showed pronounced roasted or
earthy aromas with low odor thresholds.
Figure 1. Biogeneration of acyloins using whole cells of baker’s yeast.
Figure 2. Chemoenzymatic synthesis of 5,6-dihydropyrazines.
MATERIALS AND METHODS
Materials. Dried baker’s yeast was purchased from Hefe Schweiz
(Stettfurt, Switzerland). All chemicals were from Sigma Aldrich
Chemical Co. (Buchs, Switzerland) and of analytical grade quality.
Organic solvents were distilled using a Vigreux column (1 m × 1 cm)
before their use.
Gas Chromatography (GC) Analyses. GC-olfactometry (GC-O)
analyses were performed on a Agilent 6890 Series equipped with a
splitless injector and a sniffing port. The separation of volatiles was
performed using helium as the carrier gas (1.5 mL min-1) and nitrogen
(45 kPa) as the makeup gas for the flame ionization detector (FID,
250 °C). The gas flow was divided at a point 200 mm before reaching
the detector. A part of the gas flow was deviated to the FID, while the
other part was led to a sniffing port. The sniffing port was heated to
200 °C in order to prevent condensation of the molecules and ended
into a glass funnel. For the separation of the compounds, a DB-1 or a
DB-Wax capillary column (both 30 m × 0.25 mm, film thickness 0.25
µm, J & W Scientific) was used. The temperature program for the DB-1
* To whom correspondence should be addressed. Tel: (+41)21 785 8708.
Fax: (+41)21 785 8549. E-mail: marcel-alexandre.juillerat@rdls.
nestle.com.
† Nestle´ Research Center.
‡ Universita¨t Hannover.
10.1021/jf0261809 CCC: $25.00 © 2003 American Chemical Society
Published on Web 04/15/2003

3104 J. Agric. Food Chem., Vol. 51, No. 10, 2003
Kurniadi et al.
Figure 3. GC-MS/EI spectrum of 2-pentyl-3,5-dimethyl-5,6-dihydropyrazine.
Table 1. Pool of Acyloins Produced by Baker’s Yeast
substrate
acroleinb
acyloin
yield (%)
RIa
4-hydroxy-1-penten-3-one
3-hydroxy-pent-1-en-4-one
4-hydroxy-1-hexen-3-one
3-hydroxy-1-hexen-4-one
3-hydroxy-(E)-4-hexen-2-one
3-hydroxy-(E)-4-hepten-2-one
2-hydroxy-(E)-4-hepten-3-one
3-hydroxy-(E)-4-octen-2-one
2-hydroxy-(E)-4-octen-3-one
3-hydroxy-(E)-4-nonen-2-one
2-hydroxy-(E)-4-nonen-3-one
3-hydroxy-(E)-4-decen-2-one
2-hydroxy-(E)-4-decen-3-one
2-pentanon-3-ol
3-pentanon-2-ol
2-hexanon-3-ol
3-hexanon-2-ol
2-heptanon-3-ol
3-heptanon-2-ol
2-octanon-3-ol
3-octanon-2-ol
2-nonanon-3-ol
0.5
2.5
0.2
0.3
4.9
21
14
28
31
7.2
4.5
6.1
3.7
0.5
3.9
3.8
1.6
9.1
3.8
55
19
21
9.2
4.5
1.7
1372
1381
1430
1453
1481
1586
1619
1683
1698
1776
1796
1946
1959
1344
1361
1430
1423
1560
1551
1655
1647
1768
1757
1889
1874
acroleinc
Figure 4. Suggested mechanism leading to mass fragment 124 in GC-
MS/EI spectra of 5,6-dihydropyrazines.
but-(E)-2-enalb
pent-(E)-2-enalb
using a Vigreux column at 40 °C. The extracts were analyzed by GC-
FID and GC-MS using a DB-Wax capillary column.
hex-(E)-2-enalb
hept-(E)-2-enalb
oct-(E)-2-enalb
propanalb
Chemoenzymatic Synthesis of 5,6-Dihydropyrazines from (E)-
2-Alkenals. One gram of glucose (5.6 mmol) and 140 mg of sodium
pyruvate or 155 mg of sodium 2-ketobutyric acid (1.25 mmol) were
added to 50 mL of 0.1 M citrate buffer (pH 6.0), containing 10 g of
dried baker’s yeast, 2 mM thiamine pyrophosphate, and 20 mM MgSO4.
One out of several (E)-2-alkenals (1.25 mmol) was added per assay,
and the mixture was incubated at 23 °C. After 1 h, 925 mg of 1,2-
propanediamine (12.5 mmol) was added. After it was further incubated
for 90 min, the reaction mixture was centrifuged, and an aqueous
solution containing 10 mg of trimethylpyrazine (82 µmol) was added
to the supernatant as an internal standard. The mixture was extracted
twice with diethyl ether. The combined etheral solutions were dried
over Na2SO4 and concentrated using a Vigreux column at 40 °C. The
extracts were analyzed by GC-O and GC-MS, using a DB-1 capillary
column.
butanalb
pentanalb
hexanalb
heptanalb
3-nonanon-2-ol
2-decanon-3-ol
3-decanon-2-ol
octanalb
Chemoenzymatic Synthesis of Tetrahydropyrazines from Alka-
nals. One gram of glucose (5.6 mmol) and 140 mg of sodium pyruvate
(1.25 mmol) were added to 50 mL of 0.1 M citrate buffer (pH 6.0),
containing 10 g of dried baker’s yeast, 2 mM thiamine pyrophosphate,
and 20 mM MgSO4. One out of several alkanals (1.25 mmol) was added
per assay, and the mixture was incubated at 23 °C. After 1 h, the
reaction mixture was centrifuged, and the supernatant was extracted
continuously with 150 mL of pentane/CH2Cl2 (2:1) overnight. The
organic phase was dried over Na2SO4, concentrated using a Vigreux
column at 40 °C, and adjusted to a volume of 5 mL. A 925 mg amount
of 1,2-propanediamine (12.5 mmol) was slowly added, and the mix-
ture was further incubated for 90 min. An etheral solution containing
10 mg of trimethylpyrazine (82 µmol) was added as an internal stand-
ard, and the etheral phase was washed two times with water. The
extracts were analyzed by GC-O and GC-MS, using a DB-1 capillary
column.
Chemical Synthesis of 5,6-Dihydropyrazines. The synthesis of 5,6-
dihydropyrazines was performed according to the protocol reported by
Flament and Stoll (12). To 81.4 mg of 1,2-propanediamine (1.1 mmol)
in 5 mL of ether, 1 mmol of one out of several diketones (2,3-
butanedione, 2,3-pentanedione, 2,3-hexanedione, or 3,4-hexanedione)
was slowly added. After 1 h of reaction at room temperature, the organic
phases were washed twice with one volume of water, dried over
Na2SO4, concentrated, and analyzed by GC-FID and GC-MS, using a
DB-1 capillary column.
a Linear RI on a DB-Wax column. b Pyruvate was used as the cosubstrate.
c 2-Ketobutyric acid was used as the cosubstrate.
column was 5 min isothermal at 40 °C, then raised to 260 °C at 4 °C
min-1, and kept at 260 °C for 5 min, while that for the DB-Wax column
was 5 min isothermal at 40 °C, then raised to 220 °C at 5 °C min-1
,
and kept at 220 °C for 5 min. Linear retention indices (RI) were
calculated according to van den Dool and Kratz (15). GC-MS analyses
were performed on a Finnigan MAT-8430 mass spectrometer combined
with an HP 5890 gas chromatograph using the same DB-1 and BD-
Wax columns and the same chromatographic conditions as described
above. The mass spectrometer was operated in both electron impact
mode at 70 eV and positive chemical ionization at 150 eV with
ammonia as the reagent. The interface temperature was 220 °C, and
the source temperature was set at 180 °C. All data were processed with
MassLib v 8.7 (MSP Friedli, Koeniz, Switzerland).
Biogeneration of Aliphatic Acyloins Using Whole Cells of Baker’s
Yeast. One gram of glucose (5.6 mmol) and 140 mg of sodium pyruvate
or 155 mg of sodium 2-ketobutyric acid (1.25 mmol) were added to
50 mL of 0.1 M citrate buffer (pH 6.0), containing 10 g of dried baker’s
yeast, 2 mM thiamine pyrophosphate, and 20 mM MgSO4. One out of
several alkanals or (E)-2-alkenals (1.25 mmol) was added per assay,
and the mixture was incubated at 23 °C. After 1 h, the reaction mixture
was centrifuged and the cells were discarded. An aqueous solution,
containing 145 µg of 4-hydroxy-4-methyl-pentan-2-one (1.25 µmol),
was added as an internal standard to the supernatant, and the mixture
was then extracted continuously with 150 mL pentane/CH2Cl2 (2:1)
overnight. The organic phase was dried over Na2SO4 and concentrated
Determination of the Yields and Estimation of the Odor
Threshold Values. The concentrations and yields of dihydropyrazines
and tetrahydropyrazines were determined by GC using trimethylpyrazine
as an internal standard. Approximate odor threshold values of tetrahy-
dropyrazines and dihydropyrazines were estimated by GC-O using the

Chemoenzymatic Synthesis of Pyrazine Derivatives
J. Agric. Food Chem., Vol. 51, No. 10, 2003 3105
Figure 5. GC-MS/EI spectrum of 2-ethyl-3,5-dimethyl-5,6-dihydropyrazine.
Table 2. 5,6-Dihydropyrazines Tentatively Identified on the Basis of
Their MS Data Analysis
odor
threshold
mass spectra
m/z (%)
yield
(%)c
5,6-dihydropyrazinesa odor quality (ng/L air) RIb
2-ethyl-3,5-dimethyl- roasted, nutty
0.002 1038 138 (100), 137 (50),
123 (75), 68 (19),
24
24
44
21
56 (83), 42 (70)
3-ethyl-2,5-dimethyl- roasted,
popcornlike
0.227 1031 138 (100), 137 (45),
123 (67), 82 (20),
56 (79), 42 (74)
2,3-diethyl-5-methyl- roasted
0.004 1090 152 (43), 151 (20),
137 (68), 56 (100),
42 (20)
2-propyl-3,5-dimethyl- roasted
2.5
1110 152 (75), 137 (67),
124 (89), 123 (30),
109 (36), 70 (100),
42 (94)
3-propyl-2,5-dimethyl- d
>2000 1100 152 (81), 137 (51),
124 (67), 123 (85),
109 (10), 70 (100),
42 (63)
19
23
23
24
2-butyl-3,5-dimethyl- roasted
16.8
1215 166 (48), 151 (44),
137 (57), 124 (100),
123 (69), 84 (82),
42 (47)
3-butyl-2,5-dimethyl-
d
>2000 1198 166 (50), 151 (40),
137 (48), 124 (100),
123 (80), 84 (50),
42 (68)
2-pentyl-3,5-dimethyl- roasted
3.2
1311 180 (10), 165 (28),
151 (40), 137 (52),
124 (100), 123 (42),
109 (49), 98 (80),
42 (60)
Figure 6. Suggested mechanisms for the formation of 5,6-dihydropyrazines
from R,â-unsaturated acyloins and 1,2-propanediamine.
3-pentyl-2,5-dimethyl- d
>2000 1298 180 (12), 165 (17),
22
151 (37), 137 (38),
124 (100), 123 (67),
109 (22), 98 (48),
42 (58)
well-known method described in the literature (16, 17). Three test
persons participated in the sniffing. The values listed in Tables 2 and
3 specify the odor threshold ranges at which at least two of the three
assessors agreed.
2-hexyl-3,5-dimethyl-
3-hexyl-2,5-dimethyl-
d
d
>2000 1393 194 (10), 179 (21),
24
24
27
151 (28), 137 (40),
124 (100), 123 (41),
112 (39), 42 (37)
>2000 1380 194 (12), 179 (18),
151 (22), 137 (28),
124 (100), 123 (66),
112 (40), 42 (38)
RESULTS AND DISCUSSION
Biogeneration of Acyloins. The scope and limitations of
baker’s yeast-mediated carboligation reaction were investigated.
Several reactions using aliphatic aldehydes and 2-ketocarboxylic
acids (pyruvate or 2-ketobutyric acid) as substrates were carried
out (Figure 1). After they were incubated, for all aldehydes
except but-(E)-2-enal, a pair of acyloins was identified by GC-
MS (Table 1). Where available, the molecular structures of the
acyloins were determined by comparison of their GC-MS data
with those described in the literature (18, 19). The molecular
structure of 3-hydroxy-1-penten-4-one was confirmed by com-
2-heptyl-3,5-dimethyl- d
3-heptyl-2,5-dimethyl- d
>2000 1511 208 (12), 193 (22),
151 (30), 137 (41),
126 (28), 124 (100),
123 (41), 109 (38),
42 (40)
>2000 1498 208 (11), 193 (18),
26
151 (20), 137 (42),
126 (31), 124 (100),
123 (59), 109 (22),
42 (37)
a All unknown molecules are tentatively identified on the basis of their MS data
analysis. b Linear RI on a DB-1 column. c Reaction yields were determined using
trimethylpyrazine as the internal standard. d Odorless under the experimental
conditions applied.
1
paring its analytical data (GC-MS, H NMR, 13C NMR) with
those of a synthesized reference compound (20). The structures
of 4-hydroxy-1-hexen-3-one, 3-hydroxy-(E)-4-nonen-2-one, 2-hy-
droxy-(E)-4-nonen-3-one, 3-hydroxy-(E)-4-decen-2-one, and
2-hydroxy-(E)-4-decen-3-one were tentatively suggested on the
basis of their GC-MS data. As shown in Table 1, under the
used experimental conditions, the yields of generated acyloins

3106 J. Agric. Food Chem., Vol. 51, No. 10, 2003
Kurniadi et al.
Figure 7. GC-MS/EI spectrum of 2-ethyl-3,5-dimethyltetrahydropyrazine.
acid) were incubated with whole cells of baker’s yeast. 1,2-
Propanediamine was then added to the reaction media, and the
mixtures were left under stirring at room temperature. After
extraction with diethyl ether and GC-MS analysis of the organic
phases, two isomeric forms of 5,6-dihydropyrazines were
identified for each studied combination of aldehyde and
2-ketocarboxylic acids (Figure 2). The molecular structures of
2-ethyl-3,5-, 3-ethyl-2,5-, 2-propyl-3,5-, and 3-propyl-2,5-di-
methyl-5,6-dihydropyrazines as well as that of 2,3-diethyl-5-
methyl-5,6-dihydropyrazine were determined by comparison of
their RI and GC-MS spectra with those of reference substances,
which were synthesized according to Flament and Stoll (12).
The structures of the butyl-, pentyl-, hexyl-, and heptyldimethyl-
5,6-dihydropyrazines were tentatively suggested on the basis
of their GC-MS data. The MS/EI spectra of these pyrazine
derivatives as well as those of the propyl derivatives contain a
prominent mass fragment of 124 (Figure 3). The fragment can
be explained by transfer of the γ-hydrogen from the alkyl chain
to the pyrazine nitrogen, which leads to the formation of reactive
intermediate A, which easily fragments to the allyl radical-
stabilized intermediate B (Figure 4). In agreement with this
mechanism, 2-ethyl-3,5- and 3-ethyl-2,5-dimethyl-5,6-dihydro-
pyrazine as well as 2,3-diethyl-5-methyl-5,6-dihydropyrazine
do not have a mass fragment 124, due to the missing γ-hydrogen
of the alkyl chain (Figure 5). A reaction mechanism was
proposed in order to explain the formation of the 5,6-dihydro-
pyrazines (Figure 6). It is thought that for acyloin tautomer 1
Schiff base formation (A) is followed by a double bond
migration (B), which results in the formation of a energetically
more favorable conjugated π system. The resulting molecule
has an enol function, which is in tautomeric equilibrium (C)
with a keto function. The keto group reacts with the other free
amino group of 1,2-propanediamine (D) to give 5,6-dihydro-
pyrazines. The other dihydropyrazine regioisomer could be
formed when the other amino group of 1,2-propanediamine
attacks first. However, for acyloin tautomer 2, Schiff base
formation (A′) is thought to be followed by imine/enamine
tautomerization (B′), a rearrangement (C′), a second tautomer-
ization (D′), and dihydropyrazine formation (E′). Furthermore,
it cannot be excluded that tautomerization between the acyloins
occurs. The possibility that baker’s yeast might be involved in
the proposed imine/enamine tautomerization was excluded by
the reaction of acyloins and 1,2-propanediamine in the absence
of yeast cells.
Table 3. Tetrahydropyrazines Tentatively Identified on the Basis of
Their MS Data Analysis
odor
threshold
mass spectra
m/z (%)
tetrahydropyrazinesa odor quality (ng/L air) RIb
yieldc
24
trimethyl-
roasted
135
1070 126 (100), 111 (27),
96 (33), 82 (62),
70 (34), 56 (25),
42 (93)
2-ethyl-3,5-dimethyl- bread crustlike 1.9
3-ethyl-2,5-dimethyl- cooked ricelike 70
1255 140 (100), 125 (70),
111 (53), 96 (79),
82 (43), 70 (93),
56 (82), 42 (48)
1174 140 (88), 125 (100),
111 (52), 96 (70),
82 (40), 70 (96),
56 (97), 42 (76)
21
11
10
2,3-diethyl-5-methyl- roasted
4.3
1159 154 (100), 139 (54),
126 (23), 111 (33),
96 (31), 84 (43),
70 (63), 56 (70),
42 (19)
2-propyl-3,5-dimethyl- d
3-propyl-2,5-dimethyl- d
>2000 1268 154 (90), 139 (34),
125 (100), 110 (38),
96 (42), 84 (40),
11
11
70 (72), 42 (43)
>2000 1245 154 (79), 139 (12),
125 (70), 110 (38),
96 (45), 84 (90),
70 (100), 42 (69)
a All unknown molecules are tentatively identified on the basis of their MS data
analysis. b Linear RI on a DB-1 column. c Reaction yields were determined using
trimethylpyrazine as the internal standard. d Odorless under the experimental
conditions applied.
Figure 8. Tetrahydropyrazines synthesized from acyloins.
varied drastically (from 0.2 to 55%) depending on the molecular
structure of the aldehyde substrates. The pair of C8 acyloins,
formed from hexanal and hex-(E)-2-enal, was produced with
the highest total yields of 74 and 59%, respectively. C7 and C9
acyloins were generated in moderate total yields of 11.7-35%.
However, shorter acyloins than C7 or longer than C9 were
generated with yields of less than 10%. A similar influence of
the substrate structure on the yield was observed earlier when
PDC from Zygosaccharomyces bisporus was used in an
analogous reaction (19).
The odor qualities of the 5,6-dihydropyrazines were charac-
terized by GC-O, and the odor thresholds were estimated using
the method reported in the literature (16, 17). Some of the 5,6-
dihydropyrazines had pleasant roasted odor properties (Table
2). The lowest odor thresholds (approximate range) were
determined for 2-ethyl-3,5-dimethyl-5,6-dihydropyrazine and
2,3-diethyl-5-methyl-5,6-dihydropyrazine. These values were as
Chemoenzymatic Synthesis of 5,6-Dihydropyrazines. (E)-
2-Alkenals and 2-ketocarboxylic acids (pyruvate or 2-ketobutyric

Chemoenzymatic Synthesis of Pyrazine Derivatives
J. Agric. Food Chem., Vol. 51, No. 10, 2003 3107
(4) Maga, J. A.; Sizer, C. E. Pyrazines in foods. CRC Crit. ReV.
Food Technol. 1973, 4, 39-115.
(5) Vernin, G. Chemistry of Heterocyclic Compounds in FlaVors
and Aromas; Wiley: New York, 1981.
(6) Rizzi, G. P. The biogenesis of food-related pyrazines. Food ReV.
Int. 1988, 4, 375-400.
(7) Maga, J. A. Pyrazine update. Food ReV. Int. 1992, 8, 479-558.
(8) Brophy, J. J.; Cavill, G. W. K.; Plant, W. D. Volatile constituents
of an Australian ponerine ant Rhytidoponera metallica. Insect
Biochem. 1981, 11, 307.
(9) Kosuge, T.; Zenda, H.; Tsuji, K.; Yamamoto, T.; Narita, H.
Studies on Flavor Components of Foodstuffs. Part I: Distribution
of Tetramethylpyrazine in Fermented Foodstuffs. Agric. Biol.
Chem. 1971, 35, 693-696.
(10) Hwang, H.-I.; Hartman, T. G.; Rosen, R. T.; Lech, J.; Ho, C.-T.
Formation of pyrazines from the Maillard reaction of glucose
and lysine-R-amine-15N. J. Agric. Food Chem. 1994, 42, 1000-
1004.
(11) Vitzthum, O. G.; Werkhoff, P. Newly discovered nitrogen-
containing heterocycles in coffee aroma. Z. Lebensm.-Unters.
Forsch. 1974, 156, 300-307.
(12) Flament, I.; Stoll, M. Synthesis of 3-hydroxy-2-methylpyrazines
by condensation of 1,2-propanediamine and saturated 2,3-
diketones. HelV. Chim. Acta 1967, 50, 1754-1758.
(13) Akiyama, T.; Enomoto, Y.; Shibamoto, T. A new method of
pyrazine synthesis for flavor use. J. Agric. Food Chem. 1978,
26, 1176-1179.
(14) Masuda, H.; Tanaka, M.; Akiyama, T.; Shibamoto, T. Preparation
of 5-substituted 2,3-dimethylpyrazines from the reaction of 2,3-
dimethyl-5,6-dihydropyrazines and aldehydes and ketones. J.
Agric. Food Chem. 1980, 28, 244-246.
(15) Van den Dool, H.; Kratz, P. Generalization of the retention index
system including linear temperature programmed gas-liquid
partition chromatography. J. Chromatogr. 1963, 11, 463-471.
(16) Boelens, M. H.; van Gemert, L. J. Structure-activity relationship
of natural volatile nitrogen compounds. In DeVelopments in Food
FlaVours; Birch, G. G, Lidley, M. G., Eds.; Elsevier Applied
Science: Amsterdam, 1986; pp 23-49.
(17) Ullrich, F.; Grosch, W. Identification of the most intense volatile
flavour compounds formed during autoxidation of linoleic acid.
Z. Lebensm.-Unters. Forsch. 1987, 184, 277-282.
(18) Neuser, F.; Zorn, H.; Berger, R. G. Generation of odorous
acyloins by yeast pyruvate decarboxylase and their occurrence
in sherry and soy sauce. J. Agric. Food Chem. 2000, 48, 6191-
6195.
low as those of the well-known pyrazines: 2,3-diethyl-5-meth-
ylpyrazine and 2-ethyl-3,5-dimethylpyrazine (3). Both dihydro-
pyrazines showed roasted odor characteristics and odor thresh-
olds in the range of 0.002 ng/L air. Seven 5,6-dihydropyrazines
with longer alkyl substituents showed no characteristic odor at
concentrations up to 2000 ng/L (Table 2).
Chemoenzymatic Synthesis of Tetrahydropyrazines. Satu-
rated acyloins were generated with baker’s yeast except 4-hy-
droxy-3-hexanone, which was produced according to Bel Rhlid
et al. (21). The chemical reaction of these acyloins with 1,2-
propanediamine was carried out at room temperature in diethyl
ether. After incubation, tetrahydropyrazines were identified in
the reaction mixture. These compounds were tentatively char-
acterized on the basis of their GC-MS data (Figure 7). From
the theoretical point of view, the condensation of 1,2-propanedi-
amine with each acyloin could result in two tetrahydropyrazine
isomers (Figure 8). However, in our experiments, for each
acyloin, only one peak corresponding to the tetrahydropyrazines
was identified in the reaction mixture by GC-MS analysis. This
result could be explained by the fact that either the tetrahydro-
pyrazine isomers could not be separated by GC under the
conditions applied or, in contrast to the theoretical prediction,
only one isomer was formed. As shown in Table 3, some of
these tetrahydropyrazines imparted roasted, cooked ricelike, and
bread crustlike aromatic characteristics. However, their estimated
odor thresholds were significantly higher than those of the 5,6-
dihydropyrazine counterparts.
CONCLUSIONS
The results obtained in this study show that baker’s yeast is
a versatile biocatalyst, which is capable of accepting several
aliphatic aldehydes and at least two 2-ketocarboxylic acids as
substrates for acyloins generation. The acyloins formation was
followed by chemical reaction with 1,2-propanediamine under
mild conditions and led to the production of 5,6-dihydropyra-
zines and tetrahydropyrazines with yields for some of the isomer
pairs of up to 50% (Table 2). While two of the six produced
tetrahydropyrazines were odorless under the experimental
conditions applied, four of them exhibited roasted, bread
crustlike, or cooked ricelike aroma tonalities. Some of the
synthesized 5,6-dihydropyrazines can be placed among the
pyrazine derivatives with the lowest threshold values described
in the literature (3).
(19) Neuser, F. Enzymatic formation of aroma-active R-hydrosy
ketones with yeast Zygosaccharomyces bisporus. Ph.D. Thesis,
University of Hannover, 1999.
(20) Kurniadi, T. H.; Bel Rhlid, R.; Juillerat, M. A.; Schu¨ler, M.;
Berger, R. G. Enantiogenic synthesis of (R)-(-)-3-hydroxy-1-
penten-4-one. Tetrahedron: Asymmetry 2003, 14, 363-366.
(21) Bel Rhlid, R.; Renard, M. F.; Veschambre, H. Microbial
reduction of 3,4-diketones and R-ketothioacetals. Application to
a chemoenzymatic synthesis of the exo- and endo- brevicomin
enantiomers. Bull. Soc. Chim. Fr. 1996, 133, 1011-1021.
LITERATURE CITED
(1) Gallois, A. Pyrazines in Foods. State of the Art. Sci. Aliments
1984, 4, 145-166.
(2) Seitz, E. W. Fermentation production of pyrazines and terpenoids
for flavors and fragrances. In Bioprocess Production of FlaVor,
Fragrance, and Color Ingredients; Gabelmann, A., Ed.; Wiley:
New York, 1994.
(3) Wagner, R.; Czerny, M.; Bielohradsky, J.; Grosch. W. Structure-
odour-activity relationships of alkylpyrazines. Z. Lebensm.-
Unters. Forsch. 1999, 208, 308-316.
Received for review December 2, 2002. Revised manuscript received
March 11, 2003. Accepted March 16, 2003.
JF0261809