Quantitation and Enantiomeric Ratios of Aroma Compounds Formed by an Ehrlich Degradation of l -Isoleucine in Fermented Foods

Article abstract of DOI:10.1021/acs.jafc.5b05427

The conversion of parent free amino acids into alcohols by an enzymatic deamination, decarboxylation, and reduction caused by microbial enzymes was first reported more than 100 years ago and is today known as the Ehrlich pathway. Because the chiral center at the carbon bearing the methyl group in l-isoleucine should not be prone to racemization during the reaction steps, the analysis of the enantiomeric distribution in 2-methylbutanal, 2-methylbutanol, and 2-methylbutanoic acid as well as in the compounds formed by secondary reactions, such as ethyl 2-methylbutanoate and 2-methylbutyl acetate, are an appropriate measure to follow the proposed degradation mechanism in the Ehrlich reaction. On the basis of a newly developed method for quantitation and chiral analysis, the enantiomers of the five metabolites were determined in a great number of fermented foods. Whereas 2-methylbutanol occurred as pure (S)-enantiomer in nearly all samples, a ratio of almost 1:1 of (S)- and (R)-2-methylbutanal was found. These data are not in agreement with the literature suggesting the formation of 2-methylbutanol by an enzymatic reduction of 2-methylbutanal. Also, the enantiomeric distribution in 2-methylbutanoic acid was closer to that in 2-methylbutanol than to that found in 2-methylbutanal, suggesting that also the acid is probably not formed by oxidation of the aldehyde as previously proposed. Additional model studies with (S)-2-methylbutanal did not show a racemization under the conditions of food production or during workup of the sample for volatile analysis. Therefore, the results establish that different mechanisms might be responsible for the formation of aldehydes and acids from the parent amino acids in the Ehrlich pathway.

Full text of DOI:10.1021/acs.jafc.5b05427

Article
pubs.acs.org/JAFC
Quantitation and Enantiomeric Ratios of Aroma Compounds Formed
by an Ehrlich Degradation of L‑Isoleucine in Fermented Foods
Katrin Matheis, Michael Granvogl, and Peter Schieberle*
Lehrstuhl fur Lebensmittelchemie, Department fuer Chemie, Technische Universitaet Muenchen, Lise-Meitner-Straße 34, D-85354
̈
Freising, Germany
ABSTRACT: The conversion of parent free amino acids into alcohols by an enzymatic deamination, decarboxylation, and
reduction caused by microbial enzymes was first reported more than 100 years ago and is today known as the Ehrlich pathway.
Because the chiral center at the carbon bearing the methyl group in L-isoleucine should not be prone to racemization during the
reaction steps, the analysis of the enantiomeric distribution in 2-methylbutanal, 2-methylbutanol, and 2-methylbutanoic acid as
well as in the compounds formed by secondary reactions, such as ethyl 2-methylbutanoate and 2-methylbutyl acetate, are an
appropriate measure to follow the proposed degradation mechanism in the Ehrlich reaction. On the basis of a newly developed
method for quantitation and chiral analysis, the enantiomers of the five metabolites were determined in a great number of
fermented foods. Whereas 2-methylbutanol occurred as pure (S)-enantiomer in nearly all samples, a ratio of almost 1:1 of (S)-
and (R)-2-methylbutanal was found. These data are not in agreement with the literature suggesting the formation of 2-
methylbutanol by an enzymatic reduction of 2-methylbutanal. Also, the enantiomeric distribution in 2-methylbutanoic acid was
closer to that in 2-methylbutanol than to that found in 2-methylbutanal, suggesting that also the acid is probably not formed by
oxidation of the aldehyde as previously proposed. Additional model studies with (S)-2-methylbutanal did not show a
racemization under the conditions of food production or during workup of the sample for volatile analysis. Therefore, the results
establish that different mechanisms might be responsible for the formation of aldehydes and acids from the parent amino acids in
the Ehrlich pathway.
KEYWORDS: Ehrlich pathway, aroma-active compounds, L-isoleucine, quantitation, yeast, spirits, bread, dairy products
another metabolism in Ehrlich’s and Neubauer’s experiments,
such as glucolysis.
INTRODUCTION
■
Food fermentation by yeast strains or lactic acid bacteria is
known to generate volatiles from amino acid metabolism, some
of which are well-known as fusel alcohols, such as 2- or 3-
methylbutanol. However, although the name suggests a
negative effect, these alcohols positively contribute to the
overall aroma of fermented foods, such as spirits, beer, wine,
bread, or dairy products. They are known to be formed by
amino acid degradation via a pathway first suggested by Ehrlich
a century ago.1 After isolation and characterization of the
branched-chain amino acid isoleucine, he noticed structural
similarities between the amino acid and amyl alcohol as well as
between leucine and isoamyl alcohol. By addition of either
isoleucine or leucine to fermentation mixtures with yeast,
Ehrlich detected increased levels of the respective alcohols and
proposed a “hydrating” enzyme responsible for their formation.
In 1911, Neubauer and Fromherz2 proved that from 2-amino-2-
phenylacetic acid, yeast was able to generate α-ketophenylacetic
acid and α-hydroxyphenylacetic acid as well as benzaldehyde
and phenylmethanol. A transamination to the α-keto acid was
proposed as the first step, followed by a decarboxylation to the
respective aldehyde. Finally, the aldehyde should be reduced to
the corresponding alcohol in parallel to a reduction of the α-
keto acid forming the α-hydroxy acid as byproduct. Thorne3,4
demonstrated that yeasts form all corresponding alcohols by a
degradation of the respective parent amino acid, and Gale5
showed that the active transport of amino acids across the cell
membrane is an energy-requiring process, which is provided by
The identification of the key enzymatic steps in the Ehrlich
pathway started when Sentheshanmuganathan and Elsden6
investigated the metabolism of the amino acid tyrosine to
tyrosol, and further studies7 showed that cell extracts of
Saccharomyces cerevisiae were able to transfer the amino groups
of asparagine, isoleucine, leucine, methionine, norleucine,
phenylalanine, tryptophan, and tyrosine to α-ketoglutarate.
Thus, as already assumed by Ehrlich, the first step in amino acid
catabolism is catalyzed by an aminotransferase activity. In
addition, Sentheshanmuganathan7 investigated whether the
transamination is followed by a decarboxylation or vice versa.
Because tyramine was not found as an intermediate in the
conversion of tyrosine into tyrosol, he concluded that the
transamination forming the corresponding α-keto acid must be
the first step. Consequently, the author7 proposed that tyrosol
was formed by a decarboxylase activity and an NADH-
dependent reduction of the intermediate aldehyde. Decades
̀
later, Dickinson et al.8−11 and Perpete et al.12 analyzed the
catabolism of the branched-chain amino acids isoleucine,
leucine, valine and the aromatic amino acids phenylalanine
and tryptophan as well as methionine in vivo by 13C NMR
experiments combined with gas chromatography−mass spec-
Received: November 13, 2015
Revised: December 30, 2015
Accepted: December 30, 2015
© XXXX American Chemical Society
A
DOI: 10.1021/acs.jafc.5b05427
J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry
Article
sodium sulfate. The formation of (S)-2-methylbutanal was followed by
HRGC-FID.
trometry (GC-MS). Their results showed that the amino acid
catabolism induced by yeasts included three amino transferases,
five decarboxylases, and six dehydrogenases. The exact
combination of enzymes was found to depend on the respective
amino acid, the carbon source, and the stage of growth of the
yeast with the decarboxylase possessing the highest specificity.
According to a recent review by Hazelwood et al.,13 it is now
commonly accepted that the last step in the formation of
alcohols in the Ehrlich mechanism is the reduction of the
aldehyde to the corresponding alcohol. This reduction easily
occurs, for example, when an aldehyde is offered to baker’s
yeast.14
The amino acid catabolism of branched-chain, aromatic, and
sulfur-containing amino acids by lactic acid bacteria was found
to proceed via a transamination as the first step,15 resulting in
an α-keto acid, followed by a decarboxylation to the
corresponding aldehyde as shown for Streptococcus lactis var.
maltigenes strains.16 The conversion of the aldehyde to the
corresponding alcohol by an alcohol dehydrogenase of
Streptococci was first proposed by Morgan in 1966.17
In contrast to all other proteinogenic amino acids, the
degradation products formed from L-isoleucine are all chiral.
Until the present, data on the enantiomeric distribution of the
metabolites formed in foods are scarcely available. Also, the
odor thresholds of the enantiomers of the five metabolites
formed from L-isoleucine have not been published yet. Thus,
the aim of this study was to develop a method for the
separation and quantitation of the three volatile metabolites
formed from L-isoleucine, that is, (R)- and (S)-2-methylbuta-
noic acid and (R)- and (S)-2-methylbutanol as well as (R)- and
(S)-2-methylbutanal and the respective esters derived thereof,
that is, ethyl (R)- and (S)-2-methylbutanoate as well as (R)-
and (S)-2-methylbutyl acetate. An additional challenge in
method development was the separation of these 10
metabolites from the 5 odorants generated from L-leucine.
MS-EI, m/z (%) 57 (100), 41 (94), 46 (69), 58 (54), 43 (18), 45
(17), 74 (13), 86 (12, M+), 53 (7), 59 (6).
MS-CI, m/z (%) 87 (100, [M + H]+), 69 (40), 58 (20).
Retention index (RI) on OV-1701, 731; RI on BGB-174E, 1061.
Ethyl (S)-2-Methylbutanoate. (S)-2-Methylbutanoic acid (2
mmol), ethanol (14 mmol), and sulfuric acid (100 μL) were dissolved
in diethyl ether (50 mL) and stirred for 150 min at 30 °C. After
cooling to room temperature, the reaction mixture was washed with an
aqueous solution of sodium hydrogen carbonate (3 × 50 mL; 1 mol/
L), and the organic phase was dried over anhydrous sodium sulfate.
The formation of ethyl (S)-2-methylbutanoate was followed by
HRGC-FID.
MS-EI, m/z (%) 57 (100), 102 (80), 85 (46), 74 (26), 87 (16), 73
(14), 56 (16), 55 (8), 58 (6), 103 (6).
MS-CI, m/z (%) 131 (100, [M + H]+).
RI on DB-FFAP, 1016; RI on BGB-176, 1173.
Quantitation by Stable Isotope Dilution Assays. To aliquots
of liquid food samples (1−100 mL, depending on the amounts of
analytes determined in preliminary experiments), dichloromethane
(30−300 mL) and the respective isotopically labeled internal standards
(1−10 μg; dissolved in dichloromethane) were added and stirred for
30 min at room temperature. Then, the organic layer was subjected to
high-vacuum distillation using SAFE.21 The distillate was separated
into two fractions by extraction with an aqueous sodium bicarbonate
solution (0.5 mol/L; 3 × 40 mL; pH 10) obtaining either the neutral/
basic volatiles or the acidic volatiles. The fractions were washed with
brine (0.5 mol/L; 3 × 50 mL), dried over anhydrous Na2SO4, and
concentrated to ∼100 μL using a Vigreux column (60 cm × 1 cm i.d.)
followed by microdistillation.
Solid samples (1−200 g) were frozen with liquid nitrogen and finely
ground by a commercial blender. Dichloromethane (10−300 mL) and
aliquots of the internal standards (1−10 μg; dissolved in dichloro-
methane) were added and stirred at room temperature for 90 min.
After filtration and SAFE,21 the workup procedure was continued as
described above for liquid samples.
Two-Dimensional High-Resolution Gas Chromatography−
Mass Spectrometry (HRGC/HRGC-MS). HRGC/HRGC-MS was
performed with a Trace 2000 gas chromatograph (Thermo Quest,
Mainz, Germany) coupled via a moving column stream switching
system (Thermo Quest) to a CP 3800 gas chromatograph (Varian,
Darmstadt, Germany). The column used in the first dimension was a
30 m × 0.32 mm i.d. DB-FFAP (0.25 μm film thickness) (J&W
Scientific, Waldbronn, Germany) and in the second dimension either a
30 m × 0.25 mm i.d., BGB-174E or a BGB-176, respectively (0.25 μm
film thickness) (BGB Analytik, Boeckten, Switzerland). Mass spectra
were generated using a Varian Saturn 2000 ion trap mass spectrometer
running in chemical ionization mode (MS-CI) at 70 eV using
methanol as the reagent gas. The peak areas of the analyte and the
labeled standard were determined from the mass traces of the
respective protonated molecular masses or from selected fragments,
respectively (Table 1). MS response factors were determined by
measuring defined ratios (5 + 1, 3 + 1, 1 + 1, 1 + 3, 1 + 5) of the
analyte and the corresponding stable isotopically labeled standard
(Table 1). The degradation products of L-leucine, namely, 3-
methylbutanal, 3-methylbutanol, and 3-methylbutanoic acid as well
as the esters formed therefrom, that is, ethyl 3-methylbutanoate and 3-
methylbutyl acetate, were eluted close to the respective metabolites
from L-isoleucine. The following parameters were found to separate
the respective metabolites in the first dimension on an achiral DB-
FFAP column: 2-methylbutanal and 2-methylbutanol, 35 °C for 10
min, then raised at 6 °C/min to 230 °C; ethyl 2-methylbutanoate, 2-
methylbutanoic acid, and 2-methylbutyl acetate, 40 °C for 2 min,
raised at 1 °C/min to 60 °C, and finally at 10 °C/min to 230 °C. In
the second dimension, the following chiral stationary phases and oven
parameters were used: BGB-174E, 2-methylbutanal, 35 °C for 2 min,
raised at 2 °C/min to 70 °C, and finally at 40 °C/min to 200 °C; 2-
methylbutanol, 30 °C for 30 min and finally raised at 40 °C/min to
200 °C; BGB-176: 2-methylbutanoic acid, 35 °C for 2 min, raised at 2
MATERIALS AND METHODS
Fermented Food Samples. All food samples used in the study
were purchased at local supermarkets.
■
Chemicals. The following compounds were obtained from
commercial sources: (R/S)-2-methylbutanoate (99%), ethyl 3-
methylbutanoate, 3-methylbutanal, (R/S)-2-methylbutanoic acid, (S)-
2-methylbutanoic acid, 3-methylbutanoic acid, (R/S)-2-methylbutanol,
(S)-2-methylbutanol (≥95%, sum of enantiomers), 3-methylbutanol,
3-methylbutyl acetate (≥99%), oxalyl chloride, and 1,1,1-tris-
(acetyloxy)-1,1-dihydro-1,2-benziodoxol-3-(1H)-one (Dess−Martin
periodinane) (Sigma-Aldrich, Taufkirchen, Germany); (R/S)-2-meth-
ylbutanal (Alfa Aesar, Karlsruhe, Germany); (R/S)-2-methylbutyl
acetate (98%) (TCI Europe Fine Chemicals, Eschborn, Germany);
dichloromethane, ethanol, sodium carbonate, sodium chloride, sodium
hydrogen carbonate, sodium sulfate, sodium thiosulfate, and sulfuric
acid (95−98%) (Merck, Darmstadt, Germany).
Stable Isotopically Labeled Standards. The following stable
isotopically labeled standards were prepared as previously described:
[2H2]-ethyl 2-methylbutanoate,18 [2H2]-2-methylbutanal,19 [2H9]-2-
methylbutanoic acid,14 and [2H2]-3-methylbutanol.20
Syntheses. [2H2]-(S)-2-Methylbutyl acetate was synthesized by
reduction of (S)-2-methylbutanoic acid chloride with lithium
aluminum deuteride in a first step, followed by esterification of the
formed [2H2]-(S)-2-methylbutanol with acetic acid.
(S)-2-Methylbutanal. Dess−Martin periodinane (1.4 mmol) was
added to (S)-2-methylbutanol (1.2 mmol) dissolved in dichloro-
methane (20 mL) and stirred for 90 min at room temperature. The
mixture was washed with an aqueous solution of NaHSO4 followed by
sodium thiosulfate (3 × 35 mL; 1:1:1, v/v/v) by solvent extraction in a
separating funnel, and the organic phase was dried over anhydrous
B
DOI: 10.1021/acs.jafc.5b05427
J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry
Article
Table 1. Stable Isotopically Labeled Standards, Selected
Ions, Response Factors (Rf), and GC Stationary Phases Used
in the Stable Isotope Dilution Assays
Table 2. Retention Indices (RI) of Aroma-Active L-
Isoleucine and L-Leucine Degradation Products
RI on
a
ion (m/z)
DB-
FFAP
DB-
5MS
BGB-
176
BGB-
174E
a
no.
odorant
labeling of
b
internal
column
combination
1a (R)-2-methylbutanal
1b (S)-2-methylbutanal
−
−
−
860
860
851
1039
1039
1028
1061
1067
1021
b
odorant
ethyl 2-
standard
analyte standard
Rf
b
2H3
2H2
2H9
2H2
2H2
131
134
0.83 DB-FFAP/
BGB-176
b
2
3-methylbutanal
methylbutanoate
2-methylbutanal
87
89
0.96 DB-FFAP/
BGB-174E
3a ethyl (R)-2-
methylbutanoate
1016
1016
1050
1045
1045
1049
1164
1173
1175
1087
1087
898
2-methylbutanoic
acid
103
71
112
73
0.69 DB-FFAP/
BGB-176
3b ethyl (S)-2-
methylbutanoate
c
2-methylbutanol
0.89 DB-FFAP/
BGB-174E
4
ethyl 3-methylbutanoate
2-methylbutyl
acetate
131
133
0.92 DB-FFAP/
BGB-176
5a (S)-2-methylbutyl acetate
5b (R)-2-methylbutyl acetate
1111
1111
1118
1076
1076
1074
1208
1212
1219
1114
1114
929
a
b
Ions used for quantitation (MS-CI). Response factor determined by
6
3-methylbutyl acetate
analyzing defined mixtures of the analyte and the internal standard.
c
[2H2]-3-Methylbutanol was used for the quantitation of 2-
methylbutanol.
7a (R)-2-methylbutanol
7b (S)-2-methylbutanol
1197
1197
1202
934
934
930
1227
1233
1231
1105
1109
1083
8
3-methylbutanol
°C/min to 102 °C, and finally at 40 °C/min to 200 °C; ethyl 2-
methylbutanoate and 2-methylbutyl acetate, 35 °C for 2 min, raised at
2 °C/min to 70 °C, and finally at 40 to 200 °C.
c
9a (S)-2-methylbutanoic acid
9b (R)-2-methylbutanoic acid
10 3-methylbutanoic acid
1659
1659
1659
−
−
−
1435
1454
1191
1448
1448
1091
c
c
Determination of Odor Thresholds in Air. Odor thresholds in
air were determined by the following procedure:22 after adding the
reference compound (E)-2-decenal (for 2-methylbutanoic acid, 2-
methylbutanol, ethyl 2-methylbutanoate, and 2-methylbutyl acetate)
or octanal (for 2-methylbutanal), for which the odor thresholds in air
were determined earlier, gas chromatography−olfactometry (GC-O)
was performed on BGB-176 or on BGB-174E (for 2-methylbutanal),
respectively. The solution was stepwise diluted (1:1, v/v), and GC-O
of an aliquot was performed until no odorant was detectable. The odor
thresholds in air were then calculated with the published odor
thresholds of (E)-2-decenal and octanal (2.7 ng/L and 15 ng/L,
respectively).23
Model Experiments. To check the possibility of racemization of
(S)-2-methylbutanal by keto−enol tautomerism, several model
experiments were performed using a solution of enantiopure (S)-2-
methylbutanal in phosphate buffer (KH2PO4/Na2HPO4·2H2O).
Thereby, varying parameters such as pH, temperature, and time
were applied. After treatment, the aqueous solution was extracted with
dichloromethane (3 × 30 mL), and the combined organic layers were
subjected to high-vacuum distillation (SAFE). The organic distillate
was dried over anhydrous sodium sulfate and concentrated to a
volume of ∼100 μL using a Vigreux column (60 cm × 1 cm i.d.)
followed by microdistillation prior to HRGC/HRGC-MS.
a
Odorants were consecutively numbered according to their RI on the
b
DB-FFAP column. RI could not be calculated due to solvent
c
overlapping. RI could not be calculated due to poor elution
properties.
leucine could be separated successfully (Table 2). The
separation of the two enantiomers of 2-methylbutanal (1a
and 1b) and 3-methylbutanal (2) was achieved by two-
dimensional GC-MS using a column combination of DB-FFAP
and BGB-174E (Table 2). The same combination worked for
the separation of 3-methylbutanol (8) and the enantiomers of
2-methylbutanol (7a and 7b). To separate 3-methylbutanoic
acid (10) from the enantiomers of 2-methylbutanoic acid (9a
and 9b), a combination of DB-FFAP with BGB-176 was used
(Table 2). The same column combination was found to be
appropriate for the separation of ethyl 3-methylbutanoate (4)
from ethyl (R)- and ethyl (S)-2-methylbutanoate (3a and 3b)
(Figure 1) and 3-methylbutyl acetate (6) and (S)- and (R)-2-
methylbutyl acetate (5a and 5b). The elution order of the
enantiomers was determined on the respective chiral column.
Therefore, solutions either of the racemic mixtures or of the
(S)-enantiomers were mixed and analyzed.
Odor Thresholds in Air. On the basis of the methods
developed, the odor thresholds in air were determined using a
previously published method.22 Ethyl (R)- and ethyl (S)-2-
methylbutanoate showed the lowest threshold of 0.33 ng/L
(Table 3); however, there was no difference between both
enantiomers. Small differences were found for 2-methylbutanal
and 2-methylbutanol with the respective (S)-isomer showing
the lower threshold (Table 3). The highest odor thresholds
were found for (R)-2-methylbutyl acetate (>172 ng/L) and for
its (S)-enantiomer (>187 ng/L). (S)-2-Methylbutanoic acid
(18.4 ng/L) and (R)-2-methylbutanoic acid (23.4 ng/L)
revealed similar odor thresholds to (S)-2-methylbutanol. In
summary, the thresholds between the different compounds
clearly varied, but the odor thresholds of the respective
enantiomers of each compound differed only slightly.
In addition, simultaneous distillation/extraction (SDE) according to
Likens and Nickerson24 was performed at pH 4 and 8, respectively.
After 120 min, the distillate/extract was dried over anhydrous sodium
sulfate and concentrated to a volume of ∼100 μL using a Vigreux
column (60 cm × 1 cm i.d.) followed by microdistillation prior to
HRGC/HRGC-MS.
RESULTS AND DISCUSSION
■
Method Development. First, a method for the separation
of the enantiomers of the L-isoleucine degradation products,
namely, 2-methylbutanal, 2-methylbutanol, and 2-methylbuta-
noic acid as well as the esters ethyl 2-methylbutanoate and 2-
methylbutyl acetate, was developed. In addition, the five
corresponding metabolites from L-leucine interfering during
GC separation also had to be separated. Therefore, different
parameters (stationary phase, carrier gas flow, heating rate, and
injection volume) were varied to obtain a separation of all 15
compounds. As shown by the retention indices, the 10
enantiomers from L-isoleucine and the 5 metabolites from L-
C
DOI: 10.1021/acs.jafc.5b05427
J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry
Article
Figure 1. Separation of (R)- and (S)-ethyl 2-methylbutanoate on BGB-176 and quantitation of the enantiomers in Pilsner beer.
Table 3. Odor Qualities and Odor Thresholds in Air for
Enantiomers Derived from the Amino Acid L-Isoleucine
Table 5. Concentrations and Enantiomeric Ratios of (R)-
and (S)-2-Methylbutyl Acetate in Fermented Foods
a
no.
odorant
odor quality odor threshold (ng/L)
concn (μg/L)
ratio (%)
a
1a
1b
3a
3b
5a
5b
7a
7b
9a
9b
(R)-2-methylbutanal
malty
malty
fruity
fruity
fruity
fruity
malty
malty
sweaty
sweaty
3.3
1.5
(S)-
(R)-
a
fermented food
apricot brandy
enantiomer
enantiomer
(S)
(R)
(S)-2-methylbutanal
b
b
ethyl (R)-2-methylbutanoate
ethyl (S)-2-methylbutanoate
(R)-2-methylbutyl acetate
(S)-2-methylbutyl acetate
(R)-2-methylbutanol
0.33
187
470
1220
105
765
117
135
492
111
156
202
24.3
33.4
20.9
32.1
33.1
12.5
634
78.7
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
17.7
nd
nd
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
43
0
0
0.33
bourbon whiskey
schnaps
b
>172
0
b
>187
mezcal
0
b
b
b
b
>37.2
10.5
23.4
18.4
single-malt whiskey
tequila
0
(S)-2-methylbutanol
0
(R)-2-methylbutanoic acid
(S)-2-methylbutanoic acid
Pilsner beer
0
wheat beer
0
a
wheat beer, alcohol-free
red wine
0
Odor threshold in air determined with octanal as reference
compound.23 Odor threshold in air determined with (E)-2-decenal
b
0
as reference compound.23
white wine
0
black bread (pumpernickel)
pretzel
0
Table 4. Concentrations and Enantiomeric Ratios of (R)-
and (S)-2-Methylbutanol in Fermented Foods
0
French bread
rye bread
0
0
a
concn (μg/L)
ratio (%)
sourdough wheat bread
Emmental cheese
kefir
0
(S)-
(R)-
57
0
fermented food
apricot brandy
enantiomer
enantiomer
(S)
(R)
100
100
7710
133000
15500
90000
173000
150000
8170
16000
890
nd
nd
100
100
100
100
100
100
100
100
100
100
100
77
0
0
Parmesan cheese
a
0
bourbon whiskey
schnaps
Mean values of triplicates. Concentrations in solids are given in μg/
nd
0
kg. nd, not detected.
mezcal
nd
0
single-malt whiskey
tequila
nd
0
nd
0
the concentrations of the (S)- and (R)-isomers of 2-
methylbutanol were measured, and the enantiomeric ratio
was calculated (Table 4). In all alcoholic beverages, only (S)-2-
methylbutanol occurred, which was clearly correlated with the
(S)-configuration of the methyl group in L-isoleucine. The
enantiomeric ratio of 2-methylbutanol in Dornfelder red wine
had already been examined by Frank et al., revealing 100% of
the (S)-enantiomer,25 which was in good agreement with the
data obtained in the present study.
Also in the bread samples, kefir, and Parmesan cheese, the
(S)-isomer predominated, whereas in Emmental cheese a nearly
racemic mixture of the alcohol was present. Compared to the
other foods analyzed, this cheese is produced with propionic
acid bacteria, which might follow a different metabolic pathway.
On the basis of well-established biosynthetic pathways, in
microorganisms 2-methylbutanol is enzymatically esterified
with activated acetic acid. Consequently, the ester 2-
methylbutyl acetate should appear in the foods in the same
enantiomeric ratio as the alcohol. This was confirmed for all
Pilsner beer
nd
0
wheat beer
nd
0
wheat beer, alcohol-free
red wine
nd
0
58600
18200
250
nd
0
white wine
nd
0
black bread (pumpernickel)
pretzel
75.0
11.7
nd
23
9
118
91
French bread
rye bread
5340
737
100
90
0
79.2
nd
10
0
sourdough wheat bread
Emmental cheese
kefir
978
100
59
110
74.2
807
8.7
41
11
14
6330
53.3
89
Parmesan cheese
86
a
Mean values of triplicates. Concentrations in solids are given in μg/
kg. nd, not detected.
Quantitation of the L-Isoleucine Metabolites in
Alcoholic Beverages, Bread, and Dairy Products. First,
D
DOI: 10.1021/acs.jafc.5b05427
J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry
Article
Figure 2. Metabolism of L-isoleucine ((2S,3S)-2-amino-3-methylpentanoic acid) by the Ehrlich mechanism.
Table 6. Concentrations and Enantiomeric Ratios of (R)-
and (S)-2-Methylbutanal in Fermented Foods
Table 7. Influence of pH Value, Heating Conditions, and
Treatments during Workup on the Enantiomeric Ratio of 2-
a
Methylbutanal
a
concn (μg/L)
ratio (%)
enantiomeric ratio
(%)
(S)-
(R)-
fermented food
apricot brandy
enantiomer
enantiomer
(S) (R)
expt
pH
time (min)
temp (°C)
(S)
(R)
26.3
82.5
149
18.9
42.5
71.1
80.0
130
58
66
68
62
56
74
73
64
68
69
58
65
53
52
72
51
66
66
63
42
34
32
38
44
26
27
36
32
31
42
35
47
48
28
49
33
34
37
bourbon whiskey
schnaps
1
3
4
10
10
10
10
10
10
10
10
10
20
40
60
120
120
2
80
80
99.1
99.1
99.1
97.5
91.5
89.5
56.3
50.5
89.5
83.2
77.4
79.8
98.6
85.9
97.6
93.0
0.9
0.9
2
mezcal
131
3
5
80
0.9
single-malt whiskey
tequila
163
4
6
80
2.5
204
72.1
0.79
2.02
2.73
6.11
3.08
9.50
161
5
7
80
8.5
Pilsner beer
2.06
3.41
5.78
14.3
4.02
24.4
181
6
8
80
10.5
43.7
49.5
10.5
16.8
22.6
20.2
1.4
wheat beer
7
9
80
wheat beer, alcohol-free
red wine
8
12
8
80
9
80
white wine
10
11
12
8
80
black bread (pumpernickel)
pretzel
8
80
8
80
b
French bread
rye bread
48.8
24.4
22.0
53.6
33.3
63.9
45.2
9.40
20.8
27.8
17.5
39.1
13
4
100
100
22
b
14
8
14.1
2.4
c
sourdough wheat bread
Emmental cheese
kefir
15
10
6
d
16
10
100
7.0
a
2-Methylbutanal used contained 99.1% of the (S)-enantiomer.
b
Parmesan cheese
a
Treatment was done in a simultaneous steam distillation/extraction
(SDE) device.24 Extraction at room temperature was done with an
c
Mean values of triplicates. Concentrations in solids are given in μg/
d
aqueous NaHCO3 solution (three times, each 30 s, pH 10). The
kg.
amino acid, dissolved in phosphate buffer (pH 6.0), was reacted for 10
min at 100 °C in the presence of 2-oxopropanal.
ingly neither in the alcoholic beverages nor in bread or dairy
products was pure (S)-2-methylbutanal detected. The relative
amount of the (S)-isomer was lowest in the pretzel (52%) and
highest in tequila (74%) (Table 6). These results allow the
conclusion that the respective alcohols may be formed from a
different precursor rather than by a reduction of the aldehyde.
Also, if a reductase would specifically reduce (S)-2-methyl-
butanal, a high amount of the (R)-2-methylbutanal should have
been left.
Figure 3. Possible racemization of (S)-2-methylbutanal under alkaline
conditions.
alcoholic beverages, for all bread samples, and Emmental
cheese (Table 5). On the other hand, kefir and Parmesan
cheese showed slight differences in the enantiomeric distribu-
tion compared to the respective alcohol (Tables 4 and 5).
According to the scheme representing the Ehrlich mecha-
nism (Figure 2), 2-methylbutanol should be formed from 2-
methylbutanal by an enzymatic reduction. However, surpris-
Because the methyl group in 2-methylbutanal is in the α-
position to the carbonyl group, a racemization might be
possible (Figure 3). To study this assumption, (S)-2-
E
DOI: 10.1021/acs.jafc.5b05427
J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry
Article
Table 9. Concentrations and Enantiomeric Ratios of Ethyl
(R)- and (S)-2-Methylbutanoate in Fermented Foods
a
concn (μg/L)
ratio (%)
(S)-
(R)-
fermented food
apricot brandy
enantiomer
enantiomer
(S)
(R)
177
39.2
2010
128
nd
1.65
10.6
nd
100
96
0
4
bourbon whiskey
schnaps
99
1
mezcal
100
98
0
single-malt whiskey
tequila
249
5.50
nd
2
172
100
94
0
Pilsner beer
1.11
12.8
22.2
71.6
40.8
4.11
0.29
0.34
1.32
0.86
0.39
0.24
416
0.09
1.35
2.90
nd
6
wheat beer
91
9
wheat beer, alcohol-free
red wine
88
12
0
100
100
75
white wine
nd
0
black bread (pumpernickel)
pretzel
1.37
0.10
0.22
0.51
0.37
0.13
0.16
111
25
25
39
28
30
28
41
21
75
French bread
rye bread
61
72
sourdough wheat bread
Emmental cheese
kefir
70
72
59
Figure 4. Generation of 2-methylbutanal by a Strecker degradation of
Parmesan cheese
79
L-isoleucine.
a
Mean values of triplicates. Concentrations in solids are given in μg/
kg. nd, not detected.
Table 8. Concentrations and Enantiomeric Ratios of (R) and
(S)-2-Methylbutanoic Acid in Fermented Foods
2-Methylbutanal can also be generated by a Strecker-type
degradation of L-isoleucine initiated by α-dicarbonyls (Figure
4). To test the influence of this reaction on the ratio of the
enantiomers, L-isoleucine was thermally degraded in the
presence of 2-oxopropanal in a model experiment (expt 16,
Table 7). The results showed that even under these conditions
only a small racemization occurred, suggesting that a Strecker
reaction running in parallel, for example, during food
processing, might not influence the enantiomeric ratio.
Although also the influence of different microbial strains on
the racemization of 2-methylbutanal might be possible, the
clear difference between the enantiomeric ratios in 2-
methylbutanol and 2-methylbutanal suggests a different
formation mechanism of both metabolites.
The formation of 2-methylbutanoic acid from L-isoleucine via
the Ehrlich pathway is postulated to arise either directly from a
decarboxylation of 2-oxo-3-methylpentanoic acid or by an
oxidation of 2-methylbutanal, respectively (Figure 2). Thus, the
enantiomeric distribution in 2-methylbutanoic acid was also
determined in a series of fermented foods (Table 8). In all
foods, the (S)-enantiomer clearly dominated. However, only in
red wine, Emmental cheese, and kefir was the pure (S)-isomer
found. In contrast, in both wheat beer samples, black bread, and
pretzels, also a significant portion of the (R)-isomer was
detected. The data for the Dornfelder red wine and for the
whiskey agreed with previous results,25,26 suggesting that both
pathways are obviously possible, with an oxidative decarbox-
ylation from the 2-oxo acid being followed by microorganisms
in wine, Emmental cheese, or kefir.
a
concn (μg/L)
ratio (%)
(S)-
enantiomer
(R)-
enantiomer
fermented food
apricot brandy
(S)
(R)
875
159
1800
544
485
690
265
441
439
722
415
110
349
325
161
286
3100
166
494
197
35.8
222
112
24.5
162
11.5
261
257
nd
83
89
17
11
11
17
5
bourbon whiskey
schnaps
89
mezcal
83
single-malt whiskey
tequila
95
81
19
3
Pilsner beer
97
wheat beer
69
31
37
0
wheat beer, alcohol-free
red wine
63
100
98
white wine
18.0
62.2
142
24.3
54.7
41.5
nd
2
black bread (pumpernickel)
pretzel
64
36
29
7
71
French bread
rye bread
93
75
25
6
sourdough wheat bread
Emmental cheese
kefir
94
100
100
76
0
nd
0
Parmesan cheese
a
159
24
Mean values of triplicates. Concentrations in solids are given in μg/
kg. nd, not detected.
methylbutanal was treated at different pH values (expts 1−8)
(Table 7) and for different reaction times at pH 8 (expts 9−
12). The results showed that only heating under strong alkaline
conditions (expts 7 and 8) led to complete racemization of the
aldehyde. On the other hand, workup procedures such as steam
distillation (expts 13 and 14) did not, if the pH was kept below
8. An extraction of a model solution containing (S)-2-
methylbutanal with a sodium bicarbonate solution at pH 10
(expt 15) also did not cause a racemization.
The activated (S)-2-methylbutanoic acid is commonly
esterified with ethanol in microbial fermentation, leading to
ethyl (S)-2-methylbutanoate. The ester appeared in quite high
concentrations in spirits and Parmesan cheese, but in lower
amounts in beer and wine, and the bread samples as well as
kefir and Emmental cheese showed only low concentrations
F
DOI: 10.1021/acs.jafc.5b05427
J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry
Article
and studies on the formation of key odorants during dough processing.
Z. Lebensm.-Unters. Forsch. 1995, 201, 241−248.
(Table 9). In all alcoholic beverages, except the alcohol-free
wheat beer, the amount of the (S)-isomer was >90%. Because
the enantiomeric ratio in the acid (Table 8) was not in full
agreement with that in the ester (Table 9), it can be assumed
that during mash fermentation a specific esterase could be
involved in the formation of the ester.
In conclusion, the reinvestigation of the Ehrlich pathway
established that the respective alcohols are obviously not
formed by a reduction of the respective aldehyde. Although this
reduction easily occurs in model experiments, a direct reductive
decarboxylation of the respective α-oxo acid might be more
probable.
(15) Jackson, H. W.; Morgan, M. E. Identity and origin of the malty
aroma substance from milk cultures of Streptococcus lactis var.
maltigenes. J. Dairy Sci. 1954, 37, 1316−1324.
(16) Tucker, J. S.; Morgan, M. E. Decarboxylation of α-keto acids by
Streptococcus lactis var. maltigenes. Appl. Microbiol. 1967, 15, 694−700.
(17) Morgan, M. E.; Lindsay, R. C.; Libbey, L. M.; Pereira, R. L.
Identity of additional aroma constituents in milk cultures of
Streptococcus lactis var. maltigenes. J. Dairy Sci. 1966, 49, 15−18.
(18) Guth, H.; Grosch, W. Quantitation of potent odorants of virgin
olive oil by stable isotope dilution assays. J. Am. Oil Chem. Soc. 1993,
70, 513−518.
(19) Schieberle, P.; Grosch, W. Changes in the concentrations of
potent crust odourants during storage of white bread. Flavour
Fragrance J. 1992, 7, 213−218.
(20) Guth, H.; Grosch, W. Identification of the character impact
odorants of stewed beef juice by instrumental analysis and sensory
studies. J. Agric. Food Chem. 1994, 42, 2862−2866.
AUTHOR INFORMATION
Corresponding Author
*(P.S.) Phone: +49 8161 71 2932. Fax: +49 8161 71 2970. E-
mail: peter.schieberle@ch.tum.de.
■
(21) Engel, W.; Bahr, W.; Schieberle, P. Solvent assisted flavour
evaporation − a new and versatile technique for the careful and direct
isolation of aroma compounds from complex food matrices. Eur. Food
Res. Technol. 1999, 209, 237−241.
Notes
The authors declare no competing financial interest.
REFERENCES
■
(22) Buettner, A.; Schieberle, P. Aroma properties of a homologous
series of 2,3-epoxyalkanals and trans-4,5-epoxyalk-2-enals. J. Agric. Food
Chem. 2001, 49, 3881−3884.
(1) Ehrlich, F. The chemical processes accompanying yeast
fermentation (in German). Ber. Dtsch. Chem. Ges. 1907, 40, 1027−
1047.
(23) Teranishi, R.; Buttery, R. G.; Guadagni, D. G. Odor quality and
chemical structure in fruit and vegetable flavors. Ann. N. Y. Acad. Sci.
1974, 237, 209−216.
(24) Likens, S. T.; Nickerson, G. B. Detection of certain hop oil
constituents in brewing products. Proc. Am. Soc. Brew. Chem. 1964, 5−
13.
(25) Frank, S.; Wollmann, N.; Schieberle, P.; Hofmann, T.
Reconstitution of the flavor signature of Dornfelder red wine on the
basis of the natural concentrations of its key aroma and taste
compounds. J. Agric. Food Chem. 2011, 59, 8866−8874.
(26) Poisson, L.; Schieberle, P. Characterization of the key aroma
compounds in an American bourbon whiskey by quantitative
measurements, aroma recombination, and omission studies. J. Agric.
Food Chem. 2008, 56, 5820−5826.
(2) Neubauer, O.; Fromherz, K. The decomposition of amino acids
in yeast fermentation (in German). Hoppe-Seyler’s Z. Physiol. Chem.
1910, 70, 326−350.
(3) Thorne, R. S. W. The assimilation of nitrogen from amino acids
by yeast. J. Inst. Brew. 1937, 43, 288−293.
(4) Thorne, R. S. W. The growth and fermentation of a strain of S.
cerevisiae with amino acids as nutrients. J. Inst. Brew. 1941, 47, 255−
272.
(5) Gale, E. F. The assimilation of amino-acids by bacteria. 1. The
passage of certain amino-acids across the cell wall and their
concentration in the internal environment of Streptococcus faecalis. J.
Gen. Microbiol. 1947, 1, 53−76.
(6) Sentheshanmuganathan, S.; Elsden, S. The mechanism of the
formation of tyrosol by Saccharomyces cerevisiae. Biochem. J. 1958, 69,
210−218.
(7) Sentheshanmuganathan, S. The mechanism of the formation of
higher alcohols from amino acids by Saccharomyces cerevisiae. Biochem.
J. 1960, 74, 568−576.
(8) Dickinson, J. R.; Lanterman, M. M.; Danner, D. J.; Pearson, B.
M.; Sanz, P.; Harrison, S. J.; Hewlins, M. J. E. A 13C nuclear magnetic
resonance investigation of the metabolism of leucine to isoamyl
alcohol in Saccharomyces cerevisiae. J. Biol. Chem. 1997, 272, 26871−
26878.
(9) Dickinson, J. R.; Harrison, S. J.; Hewlins, M. J. E. An investigation
of the metabolism of valine to isobutyl alcohol in Saccharomyces
cerevisiae. J. Biol. Chem. 1998, 273, 25751−25756.
(10) Dickinson, J. R.; Harrison, S. J.; Dickinson, J. A.; Hewlins, M. J.
E. An investigation of the metabolism of isoleucine to active amyl
alcohol in Saccharomyces cerevisiae. J. Biol. Chem. 2000, 275, 10937−
10942.
(11) Dickinson, J. R.; Salgado, L. E. J.; Hewlins, M. J. E. The
catabolism of amino acids to long chain and complex alcohols in
Saccharomyces cerevisiae. J. Biol. Chem. 2003, 278, 8028−8034.
̀
(12) Perpete, P.; Duthoit, O.; De Maeyer, S.; Imray, L.; Lawton, A. I.;
Stavropoulos, K. E.; Gitonga, V. W.; Hewlins, M. J. E.; Dickinson, J. R.
Methionine catabolism in Saccharomyces cerevisiae. FEMS Yeast Res.
2006, 6, 48−56.
(13) Hazelwood, L. A.; Daran, J.-M.; van Maris, A. J. A.; Pronk, J. T.;
Dickinson, J. R. The Ehrlich pathway for fusel alcohol production: a
century of research on Saccharomyces cerevisiae metabolism. Appl.
Environ. Microbiol. 2008, 74, 2259−2266.
(14) Gassenmeier, K.; Schieberle, P. Potent aromatic compounds in
the crumb of wheat bread (French-type) − influence of pre-ferments
G
DOI: 10.1021/acs.jafc.5b05427
J. Agric. Food Chem. XXXX, XXX, XXX−XXX