Alcohol & Alcoholism Vol. 35, No. 6, pp.

561568, 2000

ACETALDEHYDE PRODUCTION AND METABOLISM BY HUMAN INDIGENOUS

AND PROBIOTIC LACTOBACILLUS AND BIFIDOBACTERIUM STRAINS
T. NOSOVA, H. JOUSIMIES-SOMER1, K. JOKELAINEN, R. HEINE1 and M. SALASPURO*
Research Unit of Alcohol Diseases, University Central Hospital of Helsinki and 1Anaerobe Reference Laboratory,
National Public Health Institute, Helsinki, Finland
(Received 23 December 1999; in revised form 23 May 2000; accepted 7 June 2000)
Abstract Many human gastrointestinal facultative anaerobic and aerobic bacteria possess alcohol dehydrogenase (ADH) activity
and are therefore capable of oxidizing ethanol to acetaldehyde. We examined whether human gastrointestinal lactobacilli (three strains),
bifidobacteria (five strains) and probiotic Lactobacillus GG ATCC 53103 are also able to metabolize ethanol and acetaldehyde in vitro.
Acetaldehyde production by bacterial suspensions was determined by gas chromatography after a 1-h incubation with 22 mM ethanol.
To determine the acetaldehyde consumption, the suspensions were incubated with 50 M or 500 M acetaldehyde as well as with 500 M
acetaldehyde and 22 mM ethanol, i.e. under conditions resembling those in the human colon after alcohol intake. The influence of growth
media and bacterial concentration on the ability of lactobacilli to metabolize acetaldehyde and to produce acetate from acetaldehyde
were determined. ADH and aldehyde dehydrogenase (ALDH) activities were determined spectrophotometrically. Neither measurable
ADH nor ALDH activities were found in aerobically grown Lactobacillus GG ATCC 53103 and Lactobacillus acidophilus ATCC 4356
strains. All the lactobacilli and bifidobacteria strains revealed a very limited capacity to oxidize ethanol to acetaldehyde in vitro.
Lactobacillus GG ATCC 53103 had the highest acetaldehyde-metabolizing capacity, which increased significantly with increasing
bacterial concentrations. This was associated with a marked production of acetate from acetaldehyde. The type of the growth media had
no effect on acetaldehyde consumption. Addition of ethanol to the incubation media diminished the acetaldehyde-metabolizing capacity
of all strains. However, in the presence of ethanol, Lactobacillus GG ATCC 53103 still demonstrated the highest capacity for acetaldehyde
metabolism of all strains. These data suggest a beneficial impact of Lactobacillus GG ATCC 53103 on high gastrointestinal acetaldehyde
levels following alcohol intake. The possible clinical implications of this finding remain to be established in in vitro studies.

INTRODUCTION

colonizing the human gastrointestinal tract decreases the level

of faecal-glucuronidase (Goldin et al., 1992; Saxelin et al.,
1995) the activity of which may be associated with the
production of potentially carcinogenic substances. In addition,
feeding of Lactobacillus GG to rats has been demonstrated
to diminish the incidence of procarcinogen-induced colon
tumours (Goldin et al., 1996), and also to reduce both the
endotoxaemia and the severity of alcohol-induced experimental
liver injury in chronically ethanol-fed rats (Nanji et al., 1994).
Some lactic acid bacteria, such as group N streptococci and
Leuconostoc cremoris, are able to form small amounts of
acetaldehyde and ethanol from glucose (Lees and Jago, 1976).
However, to date, little is known about the capacity of
indigenous lactobacilli, bifidobacteria, and probiotic Lactobacillus GG ATCC 53103 to metabolize ethanol and its first
metabolite, acetaldehyde.
The aim of this study was to elucidate in vitro the capacity
of human lactobacilli and bifidobacteria to metabolize ethanol
and acetaldehyde under conditions that may prevail in the
human large intestine after moderate alcohol intake. To this
end, we examined whether aerobically grown lactobacillus
strains possess either ADH or ALDH activity. Furthermore,
we studied whether aerobically and anaerobically grown lactobacillus and anaerobically grown bifidobacterium strains can
metabolize ethanol to carcinogenic and toxic acetaldehyde
and further metabolize acetaldehyde to non-toxic compounds,
such as acetate, in vitro. In addition, we evaluated the effect of
growth media (lactobacillus selective medium MRS or nonselective brucella agar) and bacterial concentration on the
ability of lactobacilli to metabolize acetaldehyde to acetate.
Finally, we determined the ability of lactobacilli and bifidobacteria to metabolize acetaldehyde in vitro in the presence of
ethanol, i.e. under conditions more closely resembling those in
the human large intestine after alcohol intake.

Many facultative anaerobic and aerobic gastrointestinal

bacterial strains have substantial alcohol dehydrogenase
(ADH; EC 1.1.1.1) activity in vitro (Jokelainen et al., 1996a;
Nosova et al., 1997) and, correspondingly, are capable of
oxidizing ethanol to acetaldehyde in vitro and in vivo
(Jokelainen et al., 1996a,b). In addition, some facultative
and aerobic bacterial strains isolated from the human gastrointestinal tract have aldehyde dehydrogenase (ALDH, EC
1.2.1.3) activity and, therefore, are capable of metabolizing
acetaldehyde to acetate in vitro (Nosova et al., 1996). The
ALDHs of intestinal bacteria, such as members of the Enterobacteriaceae, however, have a limited capacity to metabolize
acetaldehyde produced by their ADHs (Nosova et al., 1998).
This may explain the accumulation of acetaldehyde in the
large intestine during ethanol oxidation in vitro (Jokelainen
et al., 1996b).
Microaerophilic lactobacilli and anaerobic bifidobacteria
are important members of the human indigenous flora of the
large intestine, levels up to 1010 and 1012 colony forming units
(CFU)/g respectively of intestinal contents have been recovered
in human faeces (Goldin and Gorbach, 1984). Both lactobacilli
and bifidobacteria have been suggested to have probiotic or
beneficial effects in humans and are therefore used in
probiotic dairy products (Fuller and Gibson, 1997). Some
lactobacilli (Coconnier et al., 1992), especially Lactobacillus
GG ATCC 53103 (Elo et al., 1991), and some strains of
bifidobacteria (Bernet et al., 1993) adhere in vitro to human
cell lines. It has been shown that Lactobacillus GG while
*Author to whom correspondence should be addressed at: Research Unit
of Alcohol Diseases, University Central Hospital of Helsinki, PL 345,
Tukholmankatu 8 F, 00029 HYKS, Finland.
561

2000 Medical Council on Alcoholism

562

T. NOSOVA et al.

MATERIALS AND METHODS

Culture conditions
Bacteria (Tables 1 and 2) were grown on brucella agar
plates (BBL, Cockeysville, MD, USA), supplemented with
5% defibrinated sheep blood, in an anaerobic chamber filled
with mixed gas (5% H2, 10% CO2, 85% N2) at 35C for 72 h.
In addition, Lactobacillus GG ATCC 53103 and Lactobacillus
acidophilus AHP 6509 were grown on MRS agar (Difco,
Detroit, MI, USA). These anaerobically (AN) incubated bacteria
were harvested and washed three times with pre-reduced
100 mM potassium phosphate buffer (pH 7.4) in anaerobic
conditions. Parallel cultures of each lactobacillus strain were
also incubated and processed aerobically (AE). The bifidobacterial parallel cultures were grown in anaerobic conditions,
but processed inside (AN) and outside the chamber (AE). The
bacterial suspensions were adjusted to contain 109 CFU/ml
with the aid of the McFarland standard. The actual number
of viable bacteria was determined at the beginning of the
incubation by quantitative viable count and was expressed as
CFU/ml in the vials.
ADH/ALDH activity
To study ADH and ALDH activities, aerobically grown
lactobacilli (Table 1) were sonicated for 30 s 5 in an ice bath;
supernatants were then centrifuged at 100 000 g for 65 min to
obtain cytosols which were stored at 80C until analysis. ADH
activity was determined spectrophotometrically by measurement
of the reduction of oxidized nicotinamide adenine dinucleotide (NAD+, final concentration 2.5 mM) at 340 nm at 25C in
0.1 M glycine buffer (pH 9.6) using ethanol concentrations
from 12.5 mM to 2.5 M. We have shown previously that
human gastrointestinal bacteria have ADH activity detectable
both at pH 9.6 and at pH 7.4. For most of the bacteria studied,

the ADH measured at pH 9.6 was significantly (P < 0.002)

higher than that determined at pH 7.4. Only the ADH of
Klebsiella oxytoca showed almost equal activities at both pH
levels (Nosova et al., 1997). In this study, we used pH 9.6 for
the determination of the ADH activity of the Lactobacillus
strains, as according to our previous data an alkaline pH level
may promote the determination of the ADH activity in the
bacteria. ALDH activity was determined in 60 mM sodium
pyrophosphate buffer (pH 8.8) containing NAD+ or oxidized
nicotinamide adenine dinucleotide phosphate (NADP+) (final
concentrations of either co-factors 1 mM) and 4-methylpyrazole
(final concentration 0.1 mM) added to inhibit ADH activity.
The reduction of the co-factors NAD+ or NADP+ was
monitored at 340 nm after the addition of acetaldehyde (from
0.05 to 5 mM, final concentrations). Proteins were assayed
according to Lowry et al. (1951). ADH and ALDH activities
were then calculated as nmol of reduced NAD(P)H
produced/mg protein/min. We have previously determined the
ALDH activity in the cytosolic fraction of human gastrointestinal aerobic bacteria both at pH 8.8 and at pH 7.4. We
found no significant difference between ALDH activities of the
same bacteria determined at these two pHs. (Nosova et al.,
1996). In a preliminary study, we found that Lactobacillus GG
ATCC 53103 did not have either NAD+-linked ADH- or
NADP+-linked ALDH activity at 25 or 37C. However, the
construction of the spectrophotometric device used makes it
difficult to keep a stable temperature at 37C. Because we did
not find any difference in enzymatic activities determined at
25 and 37C, we performed the enzyme assays at 25C.
Acetaldehyde production
The capacity of bacteria (Table 1) to produce acetaldehyde
from ethanol was determined by incubation of duplicated
samples containing 450 l of intact bacterial suspension in

Table 1. Bacterial strains and growth conditions used in studies on ADH/ALDH activity, acetaldehyde metabolism, and acetate production
Name
Lactobacillus GG
Lactobacillus acidophilus
Lactobacillus brevis
Lactobacillus fermentum
Bifidobacterium bifidum
Bifidobacterium dentium
Bifidobacterium spp.
Bifidobacterium spp.
Bifidobacterium spp.

Strain

Growth atmosphere and harvesting

Growth media

ATCC 53103
AHP 6509
AHP 6281
AHP 6412
ATCC 15696
ATCC 15423
AHP 16301
AHP 16455
AHP 16467

Anaerobic and aerobic

Anaerobic and aerobic
Anaerobic and aerobic
Anaerobic and aerobic
Anaerobic and aerobic
Anaerobic and aerobic
Anaerobic and aerobic
Anaerobic and aerobic
Anaerobic and aerobic

MRS and brucella agar

MRS and brucella agar
Brucella agar
Brucella agar
Brucella agar
Brucella agar
Brucella agar
Brucella agar
Brucella agar

The above strains have been used in the studies whose results are shown in Figs 14.

Table 2. Bacterial strains used in studies on acetaldehyde metabolism

Name
Lactobacillus GG
Lactobacillus brevis
Lactobacillus acidophilus
Bifidobacterium bifidum
Bifidobacterium spp.
Bifidobacterium spp.

Strain

Growth atmosphere and harvesting

Growth media

ATCC 53103
AHP 6281
AHP 6509
ATCC 15696
AHP 16467
AHP 16301

Anaerobic and aerobic

Anaerobic and aerobic
Anaerobic and aerobic
Anaerobic and aerobic
Anaerobic and aerobic
Anaerobic and aerobic

Brucella agar
Brucella agar
Brucella agar
Brucella agar
Brucella agar
Brucella agar

The above strains have been used in the studies whose results are shown in Figs 5 and 6.

ACETALDEHYDE PRODUCTION AND METABOLISM BY HUMAN BACTERIA

closed vials with 50 l of 22 mM ethanol (final concentration)

for 1 h at 37C. Samples incubated without ethanol were used
as controls and were processed simultaneously as described
above.
Acetaldehyde metabolism
The capacity of bacteria (Table 1) to metabolize acetaldehyde
was determined by incubation of duplicate samples containing
450 l of the intact bacterial suspension with 50 l of 0.05 mM
or 0.5 M acetaldehyde (final concentrations) in closed vials for
1 h at 37C. Samples incubated without ethanol were used as
controls, and acetaldehyde standards prepared by incubating
the same final concentrations of acetaldehyde (0.05 mM or
0.5 M) but without bacteria were processed simultaneously as
described above.
In addition, the capacity of bacteria (Table 2) to metabolize
acetaldehyde in vitro in the presence of ethanol was determined by incubation of duplicate samples containing 350 l
of the intact bacterial suspension with 50 l of acetaldehyde
(final concentration 0.5 mM) with or without 50 l of ethanol
(final concentration 22 mM) for 1 h at 37C. Bacterial samples
incubated without ethanol and acetaldehyde were used as
controls and were processed simultaneously. Acetaldehyde
standards, prepared by using the same final concentrations of
acetaldehyde (0.05 mM or 0.5 M) and/or ethanol (22 mM),
but without bacteria, were processed as described above. To
control for artefactual acetaldehyde production, the reaction
was stopped by addition of 50 l of 6 M perchloric acid. Ethanol
and acetaldehyde were analysed by head-space gas chromatography at 37C as described previously (Jokelainen et al.,
1996a).
Acetate production
To study acetate production, duplicate suspensions of
lactobacilli (Table 1) were incubated with 5 mM acetaldehyde
(final concentration) for 1 h at 37C. Samples incubated without
acetaldehyde were used as controls and were processed simultaneously. Acetate was then measured spectrophotometrically in
bacterial supernatants obtained by centrifugation at 1000 g for
10 min, as described previously (Nosova et al., 1996).
Calculations
The amount of acetaldehyde produced was calculated by
subtracting the amount of acetaldehyde produced by controls
from the amount of acetaldehyde of the samples incubated with
ethanol. The amount of acetaldehyde metabolized was calculated by subtracting the amount of acetaldehyde produced
by the bacterial suspension after the addition of acetaldehyde,
or acetaldehyde and ethanol, from the sum of the amount
of acetaldehyde produced by controls and the appropriate
acetaldehyde standard. The amount of acetate produced was
calculated by subtracting the amount of acetate produced by
the controls from the amount of acetate of the samples incubated
with acetaldehyde.
Statistics
Results are expressed, unless otherwise mentioned, as the
means of duplicate determination with bacteria. The statistical
significance of the difference (Fig. 4) was analysed by the
t-test using InStat software (version 2.0).

563

RESULTS
Neither NAD+-linked ADH nor NAD+- or NADP+-linked
ALDH measurable activity were found in aerobically grown
Lactobacillus GG ATCC 53103 and Lactobacillus acidophilus
ATCC 4356 strains; however, substantial amounts of cytosolic
protein were determined (data not shown).
Lactobacillus strains metabolized ethanol only weakly to
acetaldehyde. Among the lactobacillus and bifidobacterium
strains, Lactobacillus GG AN ATCC 53103 (80 nmol/109
CFU/h) had the highest acetaldehyde-producing capacity
(Fig. 1). The capacity of Lactobacillus acidophilus AHP 6509
and Lactobacillus fermentum AHP 6412 to metabolize ethanol
was negligible. In contrast, nearly all aerobically and anaerobically processed bifidobacterium strains metabolized
ethanol to acetaldehyde more actively, with Bifidobacterium
spp. AN AHP 16467 (72 nmol/109 CFU/h) showing the highest acetaldehyde-producing capacity among the bifidobacteria
(Fig. 1).
All lactobacillus and bifidobacterium strains were able to
metabolize 50 M acetaldehyde, with Lactobacillus GG AN
ATCC 53103 revealing the highest capacity for acetaldehyde
consumption (228 nmol/109 CFU/h) (Fig. 2). In addition,
this strain produced the highest amount of acetate from
acetaldehyde in vitro (Fig. 3).
The capacity of Lactobacillus GG AN ATCC 53103 to
metabolize a 500 M acetaldehyde solution was detectable at
108 CFU/ml and then increased significantly (P < 0.003), with
increasing bacterial concentrations from 108 to 109 CFU/ml
(Fig. 4). At a bacterial concentration of 109 CFU/ml,
Lactobacillus GG AN ATCC 53103 had the highest capacity
to metabolize acetaldehyde among all the strains tested. Its
acetaldehyde-metabolizing capacity was also significantly
(P < 0.03 and P < 0.003) higher than that of aerobically
processed Lactobacillus GG AE ATCC 53103 (at 108 and
109 CFU/ml, respectively) (Fig. 4).
Acetaldehyde consumption was not influenced by the type
of growth medium, MRS or brucella agar (data not shown).
All bacterial strains were able to metabolize 500 M acetaldehyde both under aerobic and anaerobic conditions (Fig. 5).
Lactobacillus GG AN ATCC 53103 showed the highest
capacity to metabolize acetaldehyde (359 nmol/109 CFU/ml),
even in the presence of 22 mM ethanol (130 nmol/109 CFU/ml).
The addition of ethanol to the acetaldehyde medium, however,
diminished the acetaldehyde-consuming capacity of all
bacterial strains (Fig. 6). Moreover, the addition of ethanol
abolished totally the acetaldehyde metabolism of Lactobacillus
brevis AHP 6281 and that of Bifidobacterium bifidum ATCC
15696 (Fig. 6).
DISCUSSION
During normal alcohol metabolism, some of the ingested
ethanol is metabolized to acetaldehyde in the colon
(Jokelainen et al., 1996b). Accordingly, after alcohol
administration, the highest acetaldehyde levels of the body are
found in the large intestine (Jokelainen et al., 1996b). Many
facultative anaerobic and aerobic gastrointestinal bacteria
reveal marked ADH (Jokelainen et al., 1996a; Nosova et al.,
1997) and catalase (Tillonen et al., 1998) activity and are

564

T. NOSOVA et al.

Fig. 1. Acetaldehyde production by Lactobacillus and Bifidobacterium strains during the incubation with ethanol.
Incubation was for 1 h with a 22 mM final concentration of ethanol at 37C. Results are expressed as means of two determinations (horizontal bars).
A scatter plot represents the data ofduplicate determinations. AN and AE, anaerobically and aerobically processed strains respectively.

Fig. 2. Acetaldehyde metabolism by bacterial strains (50 M, final concentration at the beginning of the incubation).
Acetaldehyde oxidation was followed during a 1 h incubation at 37C. Results are expressed as means of two determinations (horizontal bars).
A scatterplot represents the data ofduplicate determinations. AN and AE, anaerobically and aerobically processed strains respectively.

therefore capable of producing acetaldehyde from ethanol

in vitro under aerobic and microaerobic conditions (Salaspuro
et al., 1999). A selective reduction by the antibiotic ciprofloxacin of the number of facultative anaerobic and aerobic
intestinal bacteria in rats results in a significant decrease in
the rates of normal (Jokelainen et al., 1997) and enhanced
(Nosova et al., 1999) ethanol elimination and also inhibits
alcoholic fermentation (Visap et al., 1998). A similar
inhibition of ethanol elimination by ciprofloxacin has recently

been confirmed in man (Tillonen et al., 1999). These findings

suggest that facultative anaerobic and aerobic gastrointestinal
bacteria play a crucial role in intracolonic acetaldehyde production and/or extrahepatic ethanol elimination (Jokelainen
et al., 1996a; Salaspuro, 1997).
Acetaldehyde, a highly reactive and toxic metabolite of
ethanol, has been shown in experimental animals to be
carcinogenic (International Agency for Research on Cancer,
1985) and has been linked to several toxic effects of ethanol

ACETALDEHYDE PRODUCTION AND METABOLISM BY HUMAN BACTERIA

565

Fig. 3. Acetate production by Lactobacillus strains from acetaldehyde.

Acetate production from 5 mM acetaldehyde was followed during a 1 h incubation at 37C. Results are expressed as means of two determinations
(horizontal bars). A scatter plot represents the data of duplicate determinations. AN andAE, anaerobically and aerobically processed strains
respectively.

on organs (Lieber et al., 1980; Yokoyama et al., 1998). In

addition, acetaldehyde produced via microbial ethanol
oxidation has been proposed to be a pathogenetic factor in the
development of alcohol-associated gastrointestinal injury
(Seitz et al., 1990; Salaspuro, 1997).
Most of the ingested ethanol is absorbed during the first
hour after its intake in the stomach and small intestine. Alcohol
is evenly distributed into the water phase of the human body,
so ethanol concentrations in the large intestine are comparable
to that found in the blood (Halstedt et al., 1973; Levitt et al.,
1982). The ethanol concentration of 22 mM closely corresponds to 100 mg/dl in the blood that is useful as reference in
studies on alcohol drinking. For example, after alcohol intake
of 25% v/v by men (0.8 g/kg body wt) at 1 h after alcohol
ingestion, the ethanol level in the blood as well as in the large
intestine was 100 mg/dl (Jones and Jnsson, 1994). In rats,
after ethanol administration of a dose of 1.5 g/kg body wt,
the mean intracolonic ethanol level determined at 2 h was
20 mM and the corresponding acetaldehyde level was 480 M
(Visap et al., 1998). In this study, we used 22 mM ethanol
and 500 M acetaldehyde concentrations, corresponding to
concentrations that can be found in the large intestine of
humans after drinking alcohol.
Results of the present study show that both lactobacilli
and bifidobacteria have a very weak capacity to metabolize
ethanol to acetaldehyde in vitro at concentrations that may
prevail in the human large intestine after normal social alcohol
drinking. For example, Lactobacillus GG ATCC 53103 and
Bifidobacterium spp. AHP 16467 produced nearly 20-fold
lower acetaldehyde concentrations than did the facultative
anaerobic strain Escherichia coli IH 13369, which so far has
been shown to have the highest capacity to produce acetaldehyde from ethanol under the same conditions as used in
this study (Jokelainen et al., 1996a). However, the capacity of
lactobacilli and bifidobacteria to metabolize acetaldehyde at a

Fig. 4. Acetaldehyde metabolism by Lactobacillus GG ATCC 53103 and

Lactobacillus acidophilus AHP 4356.
Acetaldehyde (500 M, final concentration at the beginning of
incubation) was measured with the above bacteria at concentrations of
108 and 109 CFU/ml during a 1 h incubation at 37C. Results are
expressed as means SD of four determinations. AN and AE,
anaerobically and aerobically processed strains respectively. **P < 0.03
when compared with the Lactobacillus strains with 108 CFU/ml;
***P < 0.03 when compared with the Lactobacillus strains at 109CFU/ml;
and Lactobacillus GG AN ATCC53103 at 108 CFU/ml.

concentration of 50 M, close to the endogenous intracolonic

acetaldehyde level (Jokelainen et al., 1996b), was comparable
to the acetaldehyde-metabolizing capacity of some of the
facultative anaerobic and aerobic bacterial strains isolated
from the human gastrointestinal tract (Nosova et al., 1996).
For example, the acetaldehyde-metabolizing capacity of
Lactobacillus GG ATCC 53103 was about half of that of

566

T. NOSOVA et al.

Fig. 5. Acetaldehyde metabolism by bacterial strains.

Acetaldehyde (500 M, final concentration at the beginning of the incubation) during metabolism was measured a 1-h incubation at 37C. Results are
expressed as means of two determinations (horizontal bars). A scatter plot represents the data ofduplicate determinations. AN and AE, anaerobically
and aerobically processed strains, respectively.

Fig. 6. Acetaldehyde metabolism in the presence of ethanol by Lactobacillus and Bifidobacterium strains.
Acetaldehyde (500 M, final concentration at the beginning of the incubation) in the presence of 22 mM ethanol (final concentration) was during a
1 h incubation at 37C. Results are expressed as means of two determinations (horizontal bars). A scatter plot represents the data of duplicate
determinations. AN and AE, anaerobically and aerobically processed strains, respectively.

Escherichia coli IH 50763, which so far has been shown to

have the highest acetaldehyde-consuming capacity among the
facultative anaerobic and aerobic bacterial strains tested
(Nosova et al., 1996).
The capacity of lactobacillus strains to metabolize acetaldehyde to acetate was also comparable to that of the facultative
anaerobic and aerobic bacterial strains isolated from the
human gastrointestinal tract (Nosova et al., 1996). Because
acetate is an intermediate product in the metabolism of

acetaldehyde by bacterial ALDHs, its substantial production

by lactobacilli further illustrates their capacity for acetaldehyde
consumption. For example, acetate production by Lactobacillus
GG ATCC 53103 was only one-fifth of that of aerobic
Pseudomonas aeruginosa ATCC 27853, which has been
shown earlier to have the highest acetate production capacity
from acetaldehyde among the facultative anaerobic and aerobic
gastrointestinal bacteria (Nosova et al., 1996). The capacity
of Lactobacillus GG AN ATCC 53103 to metabolize

ACETALDEHYDE PRODUCTION AND METABOLISM BY HUMAN BACTERIA

acetaldehyde in vitro was detectable starting from the bacterial

concentration of 108 CFU/ml and this capacity increased significantly with rising bacterial concentrations up to 109 CFU/ml.
Recently, a 1010-CFU oral dose of Lactobacillus GG ATCC
53103 to volunteers has been shown to result in faecal recovery
of Lactobacillus GG up to 107 CFU/g faeces (Saxelin et al.,
1995). Accordingly, our data suggest that sufficient acetaldehyde consumption in vivo may require a higher total daily
oral dose of Lactobacillus GG.
Most lactobacilli are microaerophilic, which explains the
poor growth and subsequent insufficient counts of aerobically
processed lactobacilli. Bifidobacteria are anaerobic; thus, they
were not suitable for the aerobically performed study of ADH
and ALDH activity. Moreover, we did not find any measurable
ADH and ALDH activity in the supernatants of aerobically
grown lactobacilli. In addition, the acetaldehyde-consuming
capacity of anaerobically processed Lactobacillus GG ATCC
53103 was significantly higher than that of the aerobically
processed one (Fig. 4). Acetate production from acetaldehyde
by anaerobically processed Lactobacillus GG ATCC 53103 was
also four times higher than that of the aerobically processed
strain. These data suggest that, for sufficient acetaldehyde
consumption in vivo by Lactobacillus GG ATCC 53103,
anaerobic conditions such as those prevailing in the lumen of
the large intestine are required in order to promote this activity.
All lactobacillus and bifidobacterium strains were effectively able to metabolize the 500 M acetaldehyde solution, i.e.
the concentration that may occur in the large intestine after
alcohol administration (Visap et al., 1998). The addition
of ethanol to the acetaldehyde media in vitro decreased the
ability of these bacterial strains to metabolize acetaldehyde.
Under anaerobic conditions, acetaldehyde consumption by
bacteria may occur either by alcoholic fermentation or via
ALDH to acetate (Zeikus, 1980). Accordingly, because the excess of ethanol inhibited alcoholic fermentation, acetaldehyde
metabolism under this condition most probably occurred
via the ALDH pathway. In the present study, acetaldehyde
metabolism by Lactobacillus brevis AHP 6281 and Bifidobacterium bifidum ATCC 16596 was totally blocked by the
addition of ethanol, which suggests that acetaldehyde
consumption by these bacterial strains occurs predominantly
via alcoholic fermentation. Other bacterial strains were able to
use both pathways for acetaldehyde consumption. Lactobacillus
GG AN ATCC 53103 possessed the highest capacity to
metabolize acetaldehyde in the presence of ethanol as well.
In conclusion, lactobacillus and bifidobacterium strains are
weak producers of toxic and carcinogenic acetaldehyde from
ethanol, but have a relatively better capacity to remove acetaldehyde at least in vitro. Their capacity to metabolize endogenous acetaldehyde is comparable to that of the facultative
anaerobic or aerobic gastrointestinal bacteria. In the presence
of ethanol, the capacity of all bacterial strains for acetaldehyde
consumption diminishes. Anaerobically processed Lactobacillus
GG ATCC 53103 had the highest capacity to metabolize
acetaldehyde both with and without the presence of ethanol
in vitro. A positive correlation existed between rising bacterial
concentrations and an increase in the acetaldehyde-metabolizing
capacity of Lactobacillus GG AN ATCC 53103. These data
suggest that lactobacilli and bifidobacteria and, especially,
Lactobacillus GG ATCC 53103, may have a beneficial regulatory effect on the intracolonic endogenous and exogenous

567

acetaldehyde levels. It remains to be established whether these

probiotics could be useful in the prevention of acetaldehydeassociated gastrointestinal morbidity associated with excessive
alcohol intake.
Acknowledgements The work was supported financially by grants
from the Yrj Jahnsson Foundation, the Finnish-Norwegian Foundation
for Medicine, the Finnish Foundation for Alcohol Studies, the Academy of
Finland, and the Finnish Medical Foundation.

REFERENCES
Bernet, M. F., Brassart, D., Neeser, J. R. and Servin, A. (1993)
Adhesion of human bifidobacterial strains to cultured human intestinal
epithelial cells and inhibition of enteropathogencell interactions.
Applied and Environmental Microbiology 59, 41214128.
Coconnier, M., Klaenhammer, T. R., Kerneis, S., Bernet, M. F. and
Servin, A. L. (1992) Protein-mediated adhesion of Lactobacillus
acidophilus strain GG2FO4 on human enterocyte and mucus secreting
cell lines in culture. Applied and Environmental Microbiology
58, 20342039.
Elo, S., Saxelin, M. and Salminen, S. (1991) Attachment of Lactobacillus
casei strain GG to human colon carcinoma cell line Caco-2: comparison
with other dairy strains. Letters in Applied Microbiology 13, 154156.
Fuller, R. and Gibson, G. R. (1997) Modification of the intestinal
microflora using probiotics and prebiotics. Scandinavian Journal of
Gastroenterology 32, 2831.
Goldin, B. R. and Gorbach, S. L. (1984) The effect of milk and
lactobacillus feeding on human intestinal bacterial enzyme activity.
American Journal of Clinical Nutrition 39, 756761.
Goldin, B. R., Gorbach, S. L., Saxelin, M., Barakat, S., Gualtieri, L.
and Salminen, S. (1992) Survival of Lactobacillus species (Strain
GG) in human gastrointestinal tract. Digestive Diseases and Science
37, 121128.
Goldin, B. R., Gualtieri, L. J. and Moore, R.P. (1996) The effect of
Lactobacillus GG on the initiation and promotion of DMH-induced
intestinal tumors in rats. Nutrition and Cancer 25, 197204.
Halstedt, C. H., Robles, E. A. and Mezey, E. (1973) Distribution
of ethanol in the human gastrointestinal tract. American Journal of
Clinical Nutrition 26, 831834.
International Agency for Research on Cancer, Working Group on
the Evaluation of the Carcinogenic Risk of Chemicals to Humans
(1985) Allyl compounds, aldehydes, epoxides and peroxides.
In IARC Monograph 36, pp. 101132. IARC, Lyon.
Jokelainen, K., Siitonen, A., Jousimies-Somer, H., Nosova, T., Heine, R.
and Salaspuro, M. (1996a) In vitro alcohol dehydrogenase mediated
acetaldehyde production by aerobic bacteria representing the normal
colonic flora in man. Alcoholism: Clinical and Experimental
Research 20, 967972.
Jokelainen, K., Matysiak-Budnik, T., Mkisalo, H., Hckerstedt, K.
and Salaspuro, M. (1996b) High intracolonic acetaldehyde levels
produced by a bacteriocolonic pathway for ethanol oxidation in
piglets. Gut 39, 100104.
Jokelainen, K., Nosova, T., Koivisto, T., Vkevinen, S., JousimiesSomer, H., Heine, R. and Salaspuro, M. (1997) Inhibition of bacteriocolonic pathway for ethanol oxidation by ciprofloxacin in rats. Life
Sciences 61, 17551762.
Jones, A. W. and Jnsson, K. A. (1994) Between-subject and withinsubject variations in the pharmacokinetics of ethanol. British Journal
of Clinical Pharmacology 37, 427431.
Lees, G. J. and Jago, G. R. (1976) Acetaldehyde: an intermediate in the
formation of ethanol from glucose by lactic acid bacteria. Journal of
Dairy Research 43, 6373.
Levitt, M. D., Doziaki, W. and Levine, A. S. (1982) Hypothesis:
activity of colonic bacteria influences organ injury from ethanol.
Hepatology 2, 598600.
Lieber, C. S., Baraona, E., Matsuda, Y., Salaspuro, M., Hasumura, M.
and Matsuzaki, S. (1980) Hepatotoxicity of acetaldehyde. In
Biological Effects of Alcohol, Begleiter, H. ed., pp. 397411. Plenum
Press, New York.

568

T. NOSOVA et al.

Lowry, O. H., Rosebrough, N. J., Farr, A. L. and Randall R. J. (1951)

Protein measurement with the Folin phenol reagent. Journal of
Biological Chemistry 193, 265275.
Nanji, A. A., Khettry, U. and Sadrzaeh, H. (1994) Lactobacillus
feeding reduces endotoxemia and severity of experimental alcoholic
liver disease. Proceedings of the Society of Experimental Biology
and Medicine 205, 243247.
Nosova, T., Jokelainen, K., Kaihovaara, P., Jousimies-Somer, H.,
Siitonen, A., Heine, R. and Salaspuro, M. (1996) Aldehyde
dehydrogenase activity and acetate production by aerobic bacteria
representing the normal flora of human large intestine. Alcohol and
Alcoholism 31, 555564.
Nosova, T., Jousimies-Somer, H., Kaihovaara, P., Jokelainen, K.,
Heine, R. and Salaspuro, M. (1997) Characteristics of alcohol dehydrogenases of certain aerobic bacteria representing human
colonic flora. Alcoholism: Clinical and Experimental Research 21,
489494.
Nosova, T., Jokelainen, K., Kaihovaara, P., Heine, R., JousimiesSomer, H. and Salaspuro, M. (1998) Characteristics of aldehyde
dehydrogenases of certain aerobic bacteria representing human
colonic flora. Alcohol and Alcoholism 33, 273280.
Nosova, T., Jokelainen, K., Kaihovaara, P., Vkevinen, S., Rautio, M.,
Jousimies-Somer, H. and Salaspuro, M. (1999) Ciprofloxacin
administration decreases the enhanced ethanol elimination in
ethanol-fed rats. Alcohol and Alcoholism 34, 4854.
Salaspuro, M. (1997) Microbial metabolism of ethanol and acetaldehyde and clinical consequences. Addiction Biology 2, 3546.

Salaspuro, V., Nyfors, S., Heine, R., Siitonen, A., Salaspuro, M. and
Jousimies-Somer, H. (1999) Ethanol oxidation and acetaldehyde
production in vitro by human intestinal strains of Escherichia coli
under aerobic, microaerobic and anaerobic conditions. Scandinavian
Journal of Gastroenterology 3, 967973.
Saxelin, M., Pessi, T. and Salminen, S. (1995) Fecal recovery following
oral administration of Lactobacillus strain GG (ATCC 53103)
in gelatine capsules to healthy volunteers. International Journal of
Food Microbiology 25, 199203.
Seitz, H., Simanowski, U., Garzon F. et al. (1990) Possible role of
acetaldehyde in ethanol-related rectal cocarcinogenesis in the rat.
Gastroenterology 98, 406413.
Tillonen, J., Kaihovaara, P. and Salaspuro, M. (1998) Role of catalase
in in vitro acetaldehyde formation by human colonic bacteria.
Alcoholism: Clinical and Experimental Research 22, 11131119.
Tillonen, J., Homann, N., Rautio, M., Jousimies-Somer, H. and
Salaspuro, M. (1999) Ciprofloxacin decreases the rate of ethanol
elimination in humans. Gut 44, 347352.
Visap, J.-P., Jokelainen, K., Nosova, T. and Salaspuro, M. (1998)
Inhibition of intracolonic acetaldehyde production by alcoholic
fermentation in rats by ciprofloxacin. Alcoholism: Clinical and
Experimental Research 22, 11611164.
Yokoyama, A., Muramatsu, T., Ohmori, T. et al. (1998) Alcohol-related
cancers and aldehyde dehydrogenase-2 in Japanese alcoholics.
Carcinogenesis 19, 13831387.
Zeikus, J. G. (1980) Chemical and fuel production by anaerobic
bacteria. Annual Reviews in Microbiology 34, 423464.