होम Journal of Agricultural and Food Chemistry Production of higher alcohols from threonine and isoleucine in alcoholic fermentations of different...

Production of higher alcohols from threonine and isoleucine in alcoholic fermentations of different types of grain mash

, ,
यह पुस्तक आपको कितनी अच्छी लगी?
फ़ाइल की गुणवत्ता क्या है?
पुस्तक की गुणवत्ता का मूल्यांकन करने के लिए यह पुस्तक डाउनलोड करें
डाउनलोड की गई फ़ाइलों की गुणवत्ता क्या है?
खंड:
21
भाषा:
english
पत्रिका:
Journal of Agricultural and Food Chemistry
DOI:
10.1021/jf60185a003
Date:
January, 1973
फ़ाइल:
PDF, 611 KB
Conversion to is in progress
Conversion to is failed
0 comments
 

अपनी समीक्षा पोस्ट करने के लिए साइन इन करें या साइन अप करें
आप पुस्तक समीक्षा लिख सकते हैं और अपना अनुभव साझा कर सकते हैं. पढ़ूी हुई पुस्तकों के बारे में आपकी राय जानने में अन्य पाठकों को दिलचस्पी होगी. भले ही आपको किताब पसंद हो या न हो, अगर आप इसके बारे में ईमानदारी से और विस्तार से बताएँगे, तो लोग अपने लिए नई रुचिकर पुस्तकें खोज पाएँगे.
REAZIN, SCALES, ANDREASEN

Production of Higher Alcohols from Threonine and Isoleucine in Alcoholic
Fermentations of Different Types of Grain Mash
George Reazin,* Harry Scales, a n d Arthur Andreasen

Threonine- Cr-14C and isoleucine- U-i-14C were
added individually to alcoholic fermentations of
different grain mashes to establish the role of
these amino acids in the production of specific
higher alcohols. When threonine- Cr-14C was
added to malt mashes, 60% of the radioactivity of
the fusel oil was recovered in 2-methyl-1-butanol
and 30% was recovered in 1-propanol; for corn
mashes the values were 86 and 8%, respectively.
When isoleucine- UJ4C was added, the only ra-

The fermentation of grain mashes by yeast results in
the formation of various substances, called congeners,
which contribute to the characteristic flavor and odor of
distilled alcoholic beverages. The major congener group
consists of higher alcohols (fusel oil), a mixture of n-propyl (1-propanol), isobutyl (2-methyl-l-propanol), d-amyl
(2-methyl-1-butanol), and isoamyl (3-methyl-1-butanol)
alcohols. Ayrapaa ( 1 9 6 7 ~Guymon (1966), Ingraham and
Guymon (1960), Thorne (1950), and others have shown
t h a t isobutyl, d-amyl, and isoamyl alcohols are produced
from the amino acids valine. isoleucine, and leucine, respectively. The biochemical mechanisms for these transformations were demonstrated for grain fermentations
used in the production of distilled alcoholic beverages
(Reazin et al., 1970).
Of these mechanisms, t h e one for the formation of n-propyl alcohol is the least understood. Guymon.et al. (1961)
added cy amino butyric acid (AABA) to a fermentation
and found an increase in n-propyl alcohol production. (Y
ketobutyric acid (AKBA) was proposed as the metabolic
intermediate between AABA and n-propyl alcohol. Guymon tested this hypothesis by adding AABA- U-14C or
AKBA-C'-I4C to fermentations and found that radioactive
n-propyl alcohol was produced. Since AABA is not a naturally occurring substrate. it was proposed that AKBA is
produced fr; om threonine, homoserine. and/or aspartic
acid.
The role of threonine as a substrate for n-propyl alcohol
formation was investigated by Guymon et al. (1961) in experiments using three threonine-requiring mutant yeasts.
Fermentations were with resting cells in a synthetic medium containing glucose but no threonine. One strain did
not produce n-propyl alcohol; the other two produced
large amounts of this alcohol and small amounts of damyl alcohol. These results suggest a complex mechanism
for formation of n-propyl alcohol.
The d a t a to be presented were obtained from conventional fermentations of grain mashes with distillers yeast,
Saccharomyces cereuisiae. Different types of mash were
used to determine the effect of substrate composition on
the production of higher alcohols. The fermentations contained either threonine- U-14C or isoleucine- U-14C. The
higher alcohols and ethyl alcohol produced were isolated
and the amount of radioactivity present was determined.
By comparing the distribution of radioactive carbon

Joseph E . Seagram and Sons, Inc., Louisville, Kentucky 40201.
50

J. Agr. Food Chem., Vol. 21, No. 1, 1973

dioactive product was 2-methyl-1-butanol. Less
2-methyl-1-butanol was produced from malt
mashes than corn mashes, even though malt
mashes contain more amino acids; also, its radioactivity was lower. Reduction of 2 - m e t h y l - l - b ~ tanol synthesis in malt mashes is attributed to
amino acid feedback inhibition. Only threonine
formed radioactive ethanol, amounting to 7% of
the total radioactivity added.

among these products, the role of these two amino acids
in higher alcohol production was determined.
MATERIAL AND METHODS
Fermentation Conditions. The following types of mash
were used: 100% barley malt, 100% rye malt, bourbon
(60% corn, 28% rye, 12% barley malt), corn (90% corn,
10% barley malt), and corn converted with Oloclast (98%
corn, 2LTo malt, 0.04% Oloclast). Oloclast is a n (Y amyloglucosidase preparation produced by Biocon Limited, Hall
Lane, Rookery Bridge, Nr. Sandback, Cheshire, England.
Each mash contained the equivalent of 1 bushel of grain
(56 lb, as is) diluted with water to a volume of 38 gal. On
a dry basis the mashes contained approximately 166 g of
grain/l.
The malt mashes were cooked a t 145'F for 30 min. The
corn and rye in the other mashes were mixed and cooked
a t 212°F for 30 min, cooled to 145"F, and saccharified with
malt for 30 min. The mash was cooled to 86°F for yeast
inoculation and controlled a t this temperature during fermentation for 3 days. In preparing the corn mash converted with Oloclast, 1% of the malt was added before cooking a t 212°F. After cooking, the mash was cooled to 145'F
and the remaining 1% malt was added, held a t 145°F for
10 min, and then cooled for yeast inoculation. The enzyme preparation was added a t 139°F during cooling. All
mashes were adjusted with sulfuric acid to p H 5.6 before
cooking and to p H 5.0 before yeast addition.
Each mash was subdivided into 1.5-1. portions in 2-1.
sterilized Florence Flasks and inoculated with 2% v/v of
yeast grown in commercial malt extract. Ten milliliters of
a stock solution of either threonine- U-14C or isoleucineC'-14C containing approximately 10 pCi of radioactive carbon was added. The radioactive amino acids were obtained from Yew England Nuclear Corp.. Boston, Mass.
The level of radioactivity in each stock solution was determined prior to use.
Chemical Analysis a n d S e p a r a t i o n of Alcohols. Samples of each mash were analyzed before yeast inoculation.
After centrifugation, total nitrogen was determined on the
supernatant by a micro-Kjeldahl method (AOAC, 1970),
amino acid nitrogen by the ninhydrin method (Moore and
Stein, 1954), and carbohydrate, after acid hydrolysis to
reducing sugars, hy t h e Somogyi-Nelson method (Neish,
1952).
A 100-ml portion of the fermented mash was distilled
and analyzed for ethyl alcohol and fusel oil content by the
Komarowsky colorimetric procedure (AOAC, 1970). The
remaining 1.4 1. of fermented mash was distilled and a

HIGHER ALCOHOLS AXD GRAIN MASH
Table I. Relationship of Mash Composition to Ethyl Alcohol
and Fusel Oil Production

Table II. Carbon-14 Recovery in Different Type Mash
Carbon-14 recovered in alcoholic distillate

Mash composition

Threonine-’4Cb
Distillate
composition

N~~~~~~~

Carbohy-

Rd/1.51.

Barley malt
Rye malt
Bourbon
Corn
Corn and GAe

107
91
768
790
721

Isoteucine-’~ C C .

%

Rd/1.5 I.

1

1043
1411
1843
21 05
1685

Yo

~

drate

Mashtype”

Mash typea

Ethyl

Total,

Total,

g/100ml

ppmb

Free a alcohol, Fusel
amino, m1/100 oil,
ppmb

Barley malt
Rye malt
Bourbon
Corn
CornandGAC

10.5
4050
394
10.9
3920
406
12.0
797
87
12.4
692
87
11.9
554
53
aGrain composition of mash listed in text. bppm
?GA = glucoamylase.

ml

ppmb

6.39
6.39
6.92
7.19
6.85

233
202
319
338
327

11
14
19
22
8
17
aGrain composition of mash type used in text. bRadioactivity of added
threonine-9.3 pCi. Wadioactivity of added isoleucine-9.9 pCi. dR =
Radioactivity in mpCi. eGA = glucoamylase.
1
8

a

= parts per million.

crude fusel oil fraction was obtained by fractional distillation in a Podbielniak still. The fusel oil fraction was separated into individual higher alcohols by preparatory gas
chromatography, as outlined in a previous publication
(Reazin et al., 1970).
Radioactive m,aasurements were made with a liquid
scintillation spectrometer in the customary manner.

RESULTS
Fusel Oil Production in Different Mashes. The 100%
barley and 100% rye malt mashes were found to contain
more nitrogen and less carbohydrate than the bourbon,
corn, and corn converted with Oloclast mashes (Table I).
Therefore, for discussion, the five different mashes will be
considered as two groups and will be referred to as “malt”
and “corn” mashes. The alcoholic distillates prepared
from these fermentations, to which radioactive amino
acids were added, were assayed to determine the amount
of radioactivity present. The data in Table I1 show that
1%of the added threonine- UJ4C radioactivity was recovered in the alcoholic distillate from malt mashes, while
8% was recoveredl in the distillate from corn mashes. The
same relationship was found for isoleucine-U-14C, i.e.,
about 1 2 and 20% recovery, respectively.

The rate of fusel oil and ethanol production during fermentation was also determined. The rate of ethanol moduction is a measure of the rate of carbohydrate utilization. The data of Figure 1 show that ethanol )s produced
a t a faster rate in 100% barley malt than in bourbon fermentations. In both types of mash, fusel oil production
did not continue after ethanol formation had stopped. In
corn mashes, the rate of carbohydrate utilization is limited primarily by a relatively slower conversion of dextrins
into fermentable w e a r (Pan et al.. 1951).
In Figure 2, data-on amino acid utilization and fusel oil
production in the barley malt fermentation show that the
amino nitrogen decreased during ethanol and fusel oil production and increased thereafter. Fusel oil production
stopped, even though an excess of amino acid substrate
was present. The experiment was repeated (Figure 2) but
in this case glucose was added at 17 hr in an amount to
give 18% (w/v) after the original carbohydrate content of
the mash was nearly exhausted. The addition of glucose
caused the fusel oil to increase while the amino acid level
decreased. The addition of sugar also caused a resumption
of ethanol production. This experiment indicates that fermentable carbohydrate is necessary for fusel oil production and amino acid utilization. The accumulation of
amino acid nitrogen after ethanol and fusel oil production
had stopped was presumably caused by continued action

~

-6

-E

-5 0
0

5

-

’,’ a - A m l n o

E

-4
O

F u s e l 011

i

-a

Nifrogen

-I

z

0

w

X
0
0
_I

-3

*

J

0

c

w

i

40

I

-2

I

- 1

IO

20

30

40

50

60

IO

20

HOURS

Figure 1. Rate of fusel oil and ethyl alcohol production in malt
and bourbon maslies. 0-0, Ethyl alcohol; 0 - 0 , fusel oil
production in bourbon mash: V - - V , ethyl alcohol: V--V,

fusel oil production in

100%

barley malt mash.

30

40

50

60

HOURS

Figure 2. Effect of added glucose on fusel oil production and

amino nitrogen utilization in malt mash fermentations. Control
fermentation: 0-0, fusel oil; 0 - 0 , LY amino nitrogen. Experimental fermentation: V-V, fusel oil; V-7,N amino nitrogen.
J.

Agr. FoodChem., Vol.

21, No. 1, 1973

51

REAZIX, SCALES, ANDREASEX

Table Ill. Higher Alcohol Composition of Fusel Oils Produced
in Fermentations of Different Mashes

Mash type
Barley malt
Rye malt
Bourbon
Corn
Corn and GAa
"GA = glucoarnylase.

n-propyl,
ppm

Isobutyl,
ppm

d-Amyl,
ppm

Isoamyl,
pprn

20
27
16
21
25

79
52
a9
81
a3

37
36
85
92
84

97
a7
129
144
135

amino acid specific activity to t h a t of the various higher
alcohols. However, measurement of the amino acid composition of the different mashes is beyond the scope of the
present study.
When radioactive isoleucine was added to the mashes
(Table IV) virtually only d-amyl alcohol was produced
from it. Malt mashes produced d-amyl alcohol with a
slightly higher specific activity (Table V) than was obtained from corn mashes, which is the opposite of that
found for threonine. Although more d-amyl alcohol was
produced from isoleucine in corn mashes, t h e results
suggest t h a t a larger proportion of d-amyl alcohol was produced from nonisoleucine substrates in corn mashes than
in malt mashes.
Production of Ethyl Alcohol from Threonine and Isoleucine. As shown in Table VI, the ethanol from all the
threonine- U-14C-labeled mashes contained about 7 9 ~of
the total radioactivity added to the mash, except when
glucoamylase was used. Isoleucine was not transformed
into ethanol.

of the malt proteolytic onzymes. Therefore, the lower fusel
oil level in distillates from malt mash is not due to a lack
of amino acids or the inability of the yeast to produce
fusel oil, but must be due in large part to a depletion of
fermentable carbohydrates.
Since the total amount of higher alcohols produced in
malt and corn mashes differed (Table I), it was of interest
to determine if there were differences in composition.
Table I11 shows that the production of n-propyl and isobutyl alcohols was similar in all mashes. However, significantly less d-amyl and isoamyl alcohols were produced in
the malt mash fermentations. Therefore, the type of mash
used can affect t h e composition as well as the amount of
fusel oil formed.
Production of Higher Alcohols from Threonine and
Isoleucine. Experiments were conducted to determine the
role of threonine and isoleucine in t h e production of the
individual higher alcohols. When the distillates produced
from fermentations containing either of these radioactive
amino acids were analyzed for the individual higher alcohols (Table IV) most of the radioactivity was found in the
d-amyl alcohol. Considerably less was found in the isoamyl alcohol. Only threonine contributed significantly to the
production of n-propyl alcohol. Isobutyl alcohol contained
little or no carbon-14.
The radioactivity per mg of carbon (specific activity) of
the different higher alcohols was determined from the
d a t a of Tables I11 and IV. As shown in Table V, fermentations of malt mash containing radioactive threonine produced n-propyl and d-amyl alcohols with lower specific
activities t h a n comparable corn mash fermentations.
These results could be explained by differences in the specific activity of the amino acid pool of the yeast due to the
differences in the amino nitrogen content of the malt and
corn mashes. However, the difference in specific activity
between the two mashes is less for n-propyl alcohol than
for d-amyl alcohol, even though the same amount of radioactive threonine was present. This difference in specific activity suggests t h a t these alcohols are being produced
from threonine by different metabolic systems. Proof of
this could be obtained by determining the levels of the individual amino acids in each mash and comparing the

DISCUSSION

Biochemical Mechanisms. The d a t a presented indicate
that n-propyl, d-amyl, and isoamyl alcohols are produced
from threonine. The formation of n-propyl and d-amyl alcohols is in agreement with data of other workers (Ayrapaa, 1968; Guymon, 1966; Guymon et al., 1961). The various biochemical reactions involved in threonine metabolism and their relationship to isoleucine and leucine metabolism are presented in Figure 3. Pathway A represents
the reactions known to be involved in the transformation
of threonine into n-propyl alcohol. Threonine is transformed
by
biodegradative threonine
dehydratase
(Greenburg, 1969) into N amino-2-butenoic acid, which
upon deamination produces the keto acid, AKBA. The
AKBA is decarboxylated and the resultant aldehyde is reduced, producing n-propyl alcohol.
The transformation of threonine into d-amyl alcohol
may be explained by Pathway B. The initial step is the
production of active cy amino-2-butenoic acid (Greenburg,
1969; Umbarger and Brown, 1957). The enzyme responsible for this reaction is termed biosynthetic threonine
dehydratase (Greenburg, 1969). It was also reported
(Greenburg, 1969) that although the threonine dehydratase enzymes of Pathways A and B are similar, the products exist in different biochemical states so t h a t the metabolic intermediates produced by one enzyme system cannot be used in the other.
The active amino-2-butenoic acid of Pathway B is
deaminated to produce active AKBA, which reacts with
acetyl coenzyme A to produce acetohydroxybutyl coenzyme A. This is a metabolic intermediate which is also
part of the metabolic system that converts carbohydrates
into d-amyl alcohol and isoleucine (Reazin e t al., 1970).
The keto acid analog of isoleucine, N keto p methylvaleric
acid (AKBMVA), may be the metabolic intermediate

Table I V . Distribution of Radioactivity in Fusel Oils Produced from T h r e ~ n i n e - ~and
~C Is~leucine-~~C
Radioactivity in higher alcohols
n-Propyl

Barley malt
Rye malt
Bourbon
Corn
Corn and GAC

28
33
58
64
53

Isobutyl

1

a
1
1
1

1
1
3
4
2

d-Amyl

0
0
0
0
0

66
52
669
667
635

Isoamyl

940
1205
I 784
201 5
1604

10
6
39
56
32

101
133

55

aa
ai

"T = fermentations with 9.3 pCi of threonine-14C/1.5 I. of mash. b I = fermentations with 9.9 pCi of i~oleucine-'~C/1.5
I. of mash. CGA = glucoamylase.

52

J. Agr. FoodChem., Vol. 21, No. 1, 1973

HIGHER ALCOHOLS AND GRAIN MASH

Table V I . Ethyl Alcohol Production from Threonine-14C and
I~oleucine-~~C

Table V. Specific Activities of Higher Alcohols Produced from
Threonine-l4C and Isoleucine-14C

Ethyl alcohol radioactivity

Specific activitya of higher alcohols
,?-Propyl
-__
Mash type

Tb

Barley malt
Rye malt
Bourbon
Corn
Corn andGAd

'1.6
'1.3

4.0
3.4
2.4

IC

Isobutyl
Tb

0 0 0
0 0 1
0 0 0
0 0 0
0 0 0

IC

1.8
1.4
7.7
, _7.1
7.4

d-Amyl
Tb

Threoninea

Isoamyl

IC

25

0.1

33
21
21
19

0.1
0.3
0.4

0.2

Tb

IC

1.0
1.5
0.4
0.6
0.6

%pecific activity = mpCi/mgC. "T = fermentations with threonined G A = glucoamylase.
l4C. = fermentation:; with is~leucine-'~C.

linking the transformation of threonine and isoleucine into
a!-amyl alcohol.
In Pathway C, threonine dehydrogenase (Green and Elliott, 1964) oxidizes the hydroxyl group of threonine to
form 2-amino-3-ketobutyric acid. This substance is unstable and a decarboxylation occurs, producing amino acetone. The amino acetone is next converted into methyl
glyoxal, which may be oxidized into pyruvic acid. The pyruvic acid may be transformed into ethyl alcohol or, by
combination with acetyl coenzyme A, into a acetolactate,
a n intermediate in t h e metabolic system for the formation
of leucine and isoamyl alcohol from carbohydrates. The
presence of this system may explain how some of the threonine- C-14C is transformed into isoamyl alcohol.
Another enzyme t h a t acts upon threonine is threonine
aldolase. In Pathway D, this enzyme transforms threonine
into glycine and acetaldehyde (Dainty, 1967) and the acetaldehyde can be reduced to ethyl alcohol. The existence
of this system could explain the transformation of threonine- C-14C into ethyl alcohol.
The d a t a obtained using isoleucine- L'-I4C showed t h a t
a!-amyl alcohol was the primary radioactive product. This
was expected sinci: this amino acid, as shown in Figure 3,
is transformed into d-amyl alcohol cia AKBMVA. Table
IV shows t h a t some radioactive isoamyl alcohol was
formed when isoleucine- L'-'4C was added to t h e mash.
Isoleucine- C-14C inay contain small amounts of leucineUJ*C, which could be responsible for the formation of radioactive isoamyl alcohol.
The existence of Pathways A and B could explain the
observation of Guymon e t a1 (1961). In the two threoninedeficient mutants t h a t produced more n-propyl alcohol
than d-amyl alcohol, Pathway A would be the most active. The mutant t h a t did not produce either alcohol
lacked both systernls.
Effect of Different Types of Mash. A smaller amount of
fusel oil is produced in malt t h a n in corn mashes. This is
explained by a depression of fusel oil production due to a
shorter fermentation period, i e . , depletion of carbohydrates andjor to a high amino nitrogen content.
Malt mash contains more protein and proteolytic enzymes than the corn mash, resulting in a larger amino
acid pool in the malt fermentations. This inhibits the
metabolic systems producing higher alcohols from carbohydrate and threonine. In corn mash fermentation the
amino acid pool is exhausted in the early stages of fermentation (Reazin e t a l , 1970) and most of the fusel oil is
produced in t h e absence of a n amino acid pool. This is in
agreement with Ayrapaa (1971), who observed that the
longer a fermentation proceeds in the absence of nitrogen,
the higher the yield of fusel oil.
The amount of fusel oil produced during a fermentation
is reported to be governed by a keto acid overflow mechanism (Ayrapaa, 1971; Guymon e t a2, 1961; Lewis, 1964).
T h a t is, t h e amount of keto acid produced by t h e yeast

RC

%

R C

%

686
553
735
51 2
382

7

11

0.1

6
8
6
4

8
8
8

0.1
0.1
0.1

Mash type
Barley malt
Rye malt
Bourbon
Corn
Corn and GAd

lsoleucineb

aRadioactive threonine added = 9.3 pCi/1.5 I. of mash. bRadioactive
isoleucine added = 9.9 fiCi/1.5 I. of mash. CR = radioactivity in mpCi/
1.5 I. of distillate. d G A = glucoamylase.

varies with t h e nitrogen concentration of the mash, and
changes in this ratio govern the amount of fusel oil produced. Thus, in corn mashes low in nitrogen, an excess of
keto acids is present and fusel oil production is high.
while in malt mashes the high nitrogen level not only lowers the amount of keto acid produced from carbohydrates
and certain amino acids, but more of the keto acid will
probably be transformed into amino acids and cell protein. The result will be a lowering in the amount of fusel
oil produced.
The changes in higher alcohol specific activity caused
by the fermentation of different types of mash can also be
explained by t h e keto acid overflow theory. If, in the production of d-amyl alcohol, the conversion rate of isoleucine into the yeast metabolic intermediate AKBMVA is
assumed to remain constant, any increase in the amount
of this intermediate produced will be due to its formation
from substrates other than isoleucine. Therefore, in malt
mash fermentations, the high level of nitrogen will cause
AKBMVA to be produced mostly from isoleucine. and,the
specific activity of the d-amyl alcohol will be higher than
if corn mash with a lower nitrogen level had been used.
The d a t a indicate t h a t this occurred, which is consistent
with the overflow theory.
Although the relationship of specific activity to mash
type observed for isoleucine differed from that observed
when threonine was used, t h e d a t a obtained using radioactive threonine are also consistent with the overflow theory.
Threonine is produced from pyruvic acid and not its keto
acid isomer, as is isoleucine. Therefore, the high nitrogen
level of the malt mash will cause some of the pyruvic acid
normally transformed into AKBMVA or AKICA to be

n
Ethyl

Alcohol

1

[Carbohydrate

THREONIkE

1

I

,

System1

Mesh Carbohydrates

forming

lAlcohol/

Figure 3. Biochemical relationships of threonine and isoleucine
in higher alcohol production.

J. Agi. FoodChem., Vol. 21, No. 1, 1973

53

DWIVEDI. ARNOLD

shunted into aspartic acid and subsequently into threonine. As a result, the threonine-U-l4C added in the malt
fermentations will be diluted with nonradioactive threonine, causing a lower specific activity. The lower threonine specific activity will result in lower higher alcohol
specific activity. This relationship of higher alcohol specific activity was demonstrated.
When threonine- U-14C is added, the specific activity of
n-propyl and d-amyl alcohol changed, using different
types of mash. This may be due to rate differences in the
transformation of threonine into these higher alcohols.
Since most of the threonine, due to blockage of Pathway
B by high nitrogen levels of the malt mash, is transformed
uia Pathway A into AKBA, the concentration of AKBA
should be higher than in the corn fermentations, where
both systems are active. This would result in a higher
level of n-propyl alcohol in malt than in corn fermentations. Although more n-propyl alcohol is produced in the
malt fermentations, had the malt fermentations lasted as
long as the corn fermentations, this difference probably
would have been greater.
Although the overflow concept accounts for much of the
variation in fusel oil production in different mashes, there
are some observations it does not explain. For example, in
the malt fermentations, a large amino acid pool existed
throughout the fermentation period, but fusel oil was
formed only during the time of active ethanol production.
It may well be that the level of reduced diphosphopyridine nucleotide. a necessary cofactor for the transformation of keto acids into higher alcohols (SentheShanmugan-

athan, 1960), becomes limiting after the cessation of ethanol production from sugar. Although large amounts of
amino acid are present, there may not be enough energy
for the Ehrlich mechanism to function.
LITERATURE CITED
Association of Official Analytical Chemists, “Official Methods,”
Method 9.054, 150; Method 42.014,858 (1970).
Ayrapaa, T.. J . Inst. Brew. 73, 17 (1967).
Ayrapaa, T., J. Inst. Brew. 71, 169 (1968).
Ayrapaa, T., J , Inst. Brew. 77, 266 (1971).
Dainty, R.. Biochem. J. 104,46P (1967).
Green, M., Elliott, W., Biochem. J. 92, 537 (1964).
Greenburg, D., Metab. Pathways, 2967 3, 122 (1969).
Guymon, J., Ingraham, J., Crowell, E!, Arch. Biochem. Biophys.
95,163 (1961).
Guymon, J., Deuelop. Ind. Microbiol. 7,88 (1966).
Ingraham, J., Guymon, J., Arch. Biochem. Biophys. 88, 157
i1960).
Lewis, ?h.,WallersteinLab. Commun. 27, 29 (1964).
Moore, S.,
Stein, W., J. Biol. Chem. 211,907 (1954).
Neish, A,, “Analytical Methods for Bacterial Fermentation,” S a tional Research Council of Canada, Prairie Regional Laboratory,
Saskatoon, Canada, 1952, pp 34-35.
Pan, S., Andreasen, A., Kolachov, P., Arch. Biochem. 30, 6
( 19: 1) .
Reazin, G., Scales. H., Andreasen, A,, J. Agr. Food Chem. 18, 585
(1970).
Thorne, R., Wallerstein Lab. Copmun. 13., 319 (1950).
SentheShanmuganathan, S., Biochem. J . 71,568 (1960).
Umbarger, E., Brown, B., J. Bacteriol. 73,105 (1957).
Received for review May 3, 1972. Accepted October 10. 1972. Presented in part at the Division of Microbial Chemistry and Technology, 162nd National Meeting of the American Chemical Society, Washington, D. C., September 1971.

Chemistry of Thiamine Degradation in Food Products and Model Systems: A Review
Basant K. Dwivedil and Roy G. Arnold*

The instability of thiamine to heat in neutral or
alkaline systems has prompted extensive study of
the chemistry of thiamine degradation. Literature dealing with the effect of p H and heat on
the thiamine molecule is reviewed. Thermal degradation products which have been reported,
such as hydrogen sulfide, elemental sulfur, 4methyl-5-(P-hydroxyethyl)thiazole,and numerous
minor products, are discussed. The extent of
thermal degradation and the nature of the products formed appear to be determined by which of
two proposed reaction mechanisms predominates,

The sensitivity of thiamine to heat and alkali was recognized almost immediately after thiamine was discovered. Considerable information about the destruction of
thiamine during cooking, processing, and storage of foods
has appeared since the isolation of thiamine by Jansen
and Donath (1926). However, most of this information is
concerned with the loss of biological activity of thiamine
during the treatment of a particular food under specified

Department of Food Science and Technology, Universit y of Nebraska, Lincoln, Nebraska 68503.
lPresent address: Department of Food Science, Cornel1
University, Ithaca, New York 14850.
54

J. Agr. Food Chem., Vol. 21, No. 1 , 1973

which is controlled by pH. Literature dealing
with the effects of other factors, including oxidation-reduction systems, inorganic bases (sulfites,
bisulfites), thiaminase enzymes, metal complexes, radiation, and ultrasonic waves, is also reviewed. Reactions of thiamine in model systems
with proteins, amino acids, carbohydrates, other
organic compounds, and certain inorganic compounds are presented. Chemical structures of thiamine degradation products reported in the literature are shown.

conditions. Only recently have the reaction products of
thiamine in foods or in model systems been identified.
Temperature, pH, and time of heating, processing, or
storage are the most important factors contributing to the
loss of thiamine in food products. Rice and Beuk (1945)
studied thiamine decomposition in pork at different temperatures. Farrer and Morrison (1949) studied thermal destruction of thiamine in buffered solutions and showed
that the rate of destruction follows the Arrhenius equation,
In k = I - E / R T , where I = constant, R = gas constant,
E = energy of activation, k = rate constant, and T =
temperature in degrees Absolute.
Subsequent studies showed that this equation can be
successfully used in predicting thiamine retention in foods