Removal of acetaldehyde from saliva by a slow-release buccal tablet of L-cysteine



High alcohol intake is an independent risk factor for upper gastrointestinal (GI)-tract cancers. There is increasing evidence that acetaldehyde, the first metabolite of ethanol, might be responsible for ethanol-associated carcinogenesis. Especially among Asian heavy drinkers with the ALDH2-deficiency gene, i.e., a genetic inability to remove acetaldehyde, the risk of digestive tract cancers is markedly increased. Local acetaldehyde production from ethanol either by oral microbes, mucosal cells or salivary glands is a plausible carcinogenic agent in the saliva. The aim of our study was to examine whether is it possible to bind carcinogenic acetaldehyde from saliva with L-cysteine, which is slowly released from a special buccal tablet. Nine healthy male volunteers took part in our study, and each subject served as his own control. A placebo or L-cysteine-containing tablet was fastened under the upper lip. Thereafter the volunteers ingested 0.8 g/kg of body weight of 10% (v/v) ethanol, and saliva samples were collected at 20 min intervals for 320 min. Salivary acetaldehyde and ethanol levels were analysed by headspace gas chromatography. The mean reduction of acetaldehyde concentration of the saliva with the L-cysteine tablet compared to placebo was 59% (CL95% 43%, 76%). Area under the curve (AUC0–320min) with the L-cysteine and placebo tablet were 54.3 ± 11 μM × hr and 162 ± 34.2 μM × hr (mean ± SEM), respectively (p = 0.003). After alcohol intake, up to two-thirds of carcinogenic acetaldehyde can be removed from saliva with a slow-releasing buccal L-cysteine drug formulation. Thus, a buccal cysteine tablet could potentially be used to prevent upper GI-tract cancers, especially among high-risk individuals. © 2001 Wiley-Liss, Inc.

Alcohol intake and tobacco smoking are the most important independent risk factors for upper digestive tract cancers.1–5 Many of the compounds in the tobacco smoke are carcinogenic, but, in contrast, the tumour-promoting effects of alcohol drinking has so far been less well defined. Acetaldehyde, the first metabolite of ethanol, has been shown to be carcinogenic in animals and there exists strong evidence of its carcinogenic action also in man.6, 7

Asian heavy drinkers with a genetically deficient aldehyde dehydrogenase (ALDH2) enzyme do show a markedly increased risk for GI-tract cancers.8 Our recent findings demonstrate that Asians with this mutant ALDH2 have 2–3 times higher salivary acetaldehyde levels after a moderate dose of ethanol than Asians with the normal ALDH2 enzyme.7 When this observation is combined with the earlier epidemiologic data, our results provide strong evidence for the local carcinogenic action of acetaldehyde produced from alcohol in the saliva by either oral microbes or in the salivary glands.

After absorption, ethanol is evenly distributed to the waterphase of the body, and ethanol concentration in the saliva equals that of the blood.9 In addition to the tissue alcohol dehydrogenase (ADH)-mediated ethanol metabolism, gastrointestinal tract bacteria and oral cavity microflora are also able to oxidize ethanol by their ADH-enzyme,10 and accordingly, particularly high concentrations of acetaldehyde are found in the saliva and in the contents of the large intestine during ethanol oxidation.11–14

Cysteine, a nonessential amino acid, has been shown to be able to react covalently with acetaldehyde to form 2-methylthiazolidine-4-carboxylic acid.15 Thereby, cysteine could prevent acetaldehyde to interact with cellular proteins and DNA, which might interfere with normal cellular functions including acetaldehyde-induced carcinogenesis.6, 7, 16, 17

The aim of our study was to examine the ability of L-cysteine, which was slowly released from a special drug formulation, to bind salivary acetaldehyde in the oral cavity after the ingestion of a dose of alcohol and thus to reduce the potential carcinogenic effects of topical acetaldehyde in the upper gastrointestinal tract.



Nine healthy male volunteers, mean age (±SEM) 27 ± 1 years, took part in our study. All of them were moderate alcohol consumers, with a weekly average consumption of 70 g or less of ethanol. One of the subjects was a light smoker (less than 7 cigarettes per day). Exclusion criteria were as follows: treatment with antibiotics or oral antiseptic in the past month, recent food or fluid intake, smoking or toothbrushing during the previous 90 min. All participants were told to refrain from ethanol for at least 36 hr before the study.

Study design

Our study was approved by the Ethical committee of the Department of Medicine, Helsinki University Central Hospital and by the Finnish National Agency for Medicines. An informed consent to participate in our study was obtained from the subjects. A paired, placebo control design in which each subject served as his own control was used. Two study days were separated by at least 48 hr. A commercially available paraffin wax chewing gum (Orion Diagnostics, Espoo, Finland) was used to stimulate the production of saliva, and the volunteers were instructed to chew at all sites of the jaw to have a representative sample. After baseline saliva collection (2 ml within 2 min), each volunteer fastened the placebo or L-cysteine-containing tablet under the upper lip at the beginning of each study and the baseline saliva collection was repeated. After the second baseline saliva collection, each volunteer ingested 0.8 g ethanol/kg body weight in a standardised 10% v/v solution of absolute ethanol in distilled water within 20 min. To remove local ethanol, the subjects rinsed their mouths with water, and thereafter the saliva samples were collected at 20 min intervals for 320 min.

L-cysteine and placebo drug formulation

The composition of the placebo tablet was 130 mg of hydroxypropyl methylcellulose (HPMC) (Methocel K4M Premium EP, Dow Chemical Co., Midland, MI), 6.9 mg carbomer (Carbopol 971P NF, BF Goodrich, Clevland, OH) and 1.4 mg magnesium stearate (Ph.Eur., Merck, Darmstadt, Germany). The composition of the L-cysteine drug formulation was exactly the same except 130 mg of HPMC was reduced to 30 mg and was replaced with 100 mg of L-cysteine (Sigma Chemical, St. Louis, MO). HPMC and carbomer molecules are widely used in oral products as a tablet binder and to retard drug release from tablet matrices.18 Regarding our study, HPMC and carbomer enables the binding of the tablet to the gingiva and furthermore the slow release of L-cysteine from the drug formulation.

In vivo salivary acetaldehyde production

To measure in vivo salivary acetaldehyde levels, 450 μl of saliva was immediately transferred into a vial that contained 50 μl of 6 mol/l perchloric acid. Acetaldehyde and ethanol levels were analysed by headspace gas chromatography as described previously.19 Each measurement was done in duplicate. To minimise the effects of artefactual acetaldehyde formation, the baseline values were subtracted from the acetaldehyde levels measured during the experiment.

Statistical analysis

All values are expressed as means ± SEM. The areas under the acetaldehyde concentration-time profiles (AUC0–320min) were determined by using NCSS 2000 statistical software (Kaysville, UT). Statistical significance of the values obtained with placebo or L-cysteine tablet was analysed by Wilcoxon signed rank test. p-values < 0.05 were regarded as significant.


In the experiments with the L-cysteine-containing tablet, the mean in vivo salivary acetaldehyde levels after ethanol ingestion were 2 to 3 times lower than during the placebo tablet experiment throughout the whole follow-up period of 320 min (Fig. 1). The area under the curve of salivary acetaldehyde with L-cysteine tablet was 54.3 ± 11 μM × hr and with placebo 162.3 ± 34.2 μM × hr (p = 0.003). Accordingly, L-cysteine reduced the exposure of the buccal cavity and esophagus to the carcinogenic acetaldehyde by mean 59 ± 8% as compared to the placebo. The corresponding in vivo salivary ethanol concentrations were equal in placebo and L-cysteine experiments during the whole follow-up period of 300 min (Fig. 2).

Figure 1.

In vivo acetaldehyde levels (mean ± SEM) in oral saliva of 9 volunteers with placebo or L-cysteine-containing buccal drug formulation after a dose of alcohol (0.8 g/kg of body weight). Differences between acetaldehyde concentrations are significant at all time points from 20 to 320 min (p ≤ 0.001).

Figure 2.

In vivo ethanol concentrations (mean ± SEM) in oral saliva of 9 volunteers with placebo or L-cysteine-containing buccal drug formulation after a dose of alcohol (0.8 g/kg of body weight).


There is extensive epidemiologic data showing that tobacco smoking and heavy alcohol consumption increase the risk of cancers of the mouth, pharynx, larynx and esophagus.1–5, 20 Both smoking and alcohol consumption have been shown to be independent risk factors for upper digestive tract cancers. When combined, there is evidence indicating that alcohol and tobacco act together in a multiplicative rather than in an additive manner, having synergistic tumour-promoting effects.5, 21 Nevertheless, the exact mechanism by which they exert their carcinogenic effect has so far been poorly understood, especially concerning alcohol consumption, because ethanol itself is not carcinogenic.22 It has been suggested that a unifying pathogenetic mechanism may underlie these epidemiologic findings. This could be the local production of carcinogenic acetaldehyde from ethanol by oral microbes, since acetaldehyde is also one of the major components in the smoke of tobacco cigarettes.23

There is increasing evidence for acetaldehyde of being the ultimate carcinogenic substance behind excessive alcohol intake. Acetaldehyde has been shown to be highly toxic and mutagenic.21, 24–27 The induction of cytogenetic effects has been postulated to be related to the ability of acetaldehyde to form DNA-DNA and/or DNA-protein cross-links.28 Moreover, there is sufficient evidence for the carcinogenicity of acetaldehyde to experimental animals.6, 29, 30 Acetaldehyde has been shown to cause inflammation and metaplasia of tracheal epithelium31, 32 and to alter the proliferation, differentiation and adhesion properties of human colon adenocarcinoma cell line Caco-2.33 Furthermore, there is significant positive correlation between ethanol-derived intracolonic acetaldehyde concentrations and hyperregenerative changes in the colorectal mucosa in rats.34

Recent epidemiologic studies have shown that the risk of ethanol-associated upper digestive tract cancers is markedly increased in heavy-drinking Asian subjects with low-activity ALDH2 enzyme. Significantly increased risks (odds ratios) in the presence of ALDH2 deficiency gene have been found for oropharyngolaryngeal (11.1), esophageal (12.5), stomach (3.5), colon (3.49) and lung (8.2), and esophageal cancer concomitant with oropharyngolaryngeal and/or stomach cancer (54.2) but not for liver or other cancers.8 Also, individuals who are homozygous for the fast alcohol-metabolising alcohol dehydrogenase (ADH3) enzyme have increased risk of alcohol-related oral cancer.35 The plausible explanation for this association is the abnormally high salivary acetaldehyde concentration after alcohol intake among individuals with ALDH2 deficiency or individuals possessing ADH3*1,1 genotype.7, 36 Hence, there is conclusive experimental evidence of the local carcinogenic action of acetaldehyde in humans.

Bacteria possessing ADH enzyme ferment acetaldehyde to ethanol in anaerobic conditions. In aerobic and microaerobic conditions, the reaction catalysed by ADH enzyme runs to the opposite direction with acetaldehyde as an end product.10, 37 This reaction leads to the formation of marked amounts of acetaldehyde in the saliva by oral microflora after the ingestion of a moderate amount of alcohol.7 Salivary acetaldehyde production can be significantly reduced by the use of an antiseptic mouthwash, underlying the microbial background for local acetaldehyde production.19 Furthermore, in vitro acetaldehyde production by mouthwashings from oropharyngolaryngeal cancer patients has been shown to be significantly higher than that of the control patients,11 and the production of acetaldehyde in saliva from ethanol is markedly increased in smokers and heavy drinkers.38

To prevent the harmful effects of ethanol-derived acetaldehyde, beside being an abstainer, we show that two-thirds of acetaldehyde can be trapped in the saliva with slowly and continuously released L-cysteine. Thiol compounds, such as cysteine, are known to be able to protect against acetaldehyde toxicity. Cysteine exerts its protective effect by forming nonenzymatically with acetaldehyde, a nonreactive, stable compound, 2-methylthiazolidine-4-carboxylic acid.15In vitro, the addition of increasing amounts of cysteine progressively decreases the detectable content of acetaldehyde.39 In animal studies, externally given thiols protect from the lethal effects of acetaldehyde.40 Pretreatment with an oral dose of L-cysteine or N-acetyl-cysteine gave 80% or more survivors in rats compared to controls when acetaldehyde (LD90 dose) was administered by oral intubation.39 Cysteine has also been reported to reverse the acetaldehyde-induced inhibition of several mitochondrial functions in vitro.41 These effects were concluded to endue from the thiol group of cysteine, which traps the reactivity of acetaldehyde.

Microbial salivary acetaldehyde production shows high interindividual variation, but there exists a highly significant positive correlation between salivary ethanol and acetaldehyde levels.38 Moreover, poor oral hygiene, heavy alcohol drinking and smoking may contribute to the individual risk of upper GI-tract cancers by increasing the acetaldehyde levels in the oral cavity.38 Our buccal drug formulation slowly releases cysteine to the oral cavity. Thus, it enables a continuous concentration of cysteine in the mouth during microbially mediated ethanol oxidation and acetaldehyde challenge. As a nonessential amino acid, L-cysteine is safe and since it markedly decreases the levels of carcinogenic acetaldehyde in the saliva, the new drug formulation is a promising new agent for the prevention of upper GI-tract cancers.

In conclusion, we have demonstrated that acetaldehyde produced in the oral cavity by microbial ethanol oxidation is significantly decreased by a new buccal L-cysteine-releasing drug formulation. L-cysteine efficiently binds reactive acetaldehyde by forming a stable, nontoxic 2-methylthiatzolidine-4-carboxylic acid compound and thus prevents acetaldehyde from interacting with cellular proteins. Hence, buccal cysteine tablets could potentially be used to prevent upper GI-tract cancers induced by ethanol-derived acetaldehyde, especially among individuals with increased acetaldehyde production in the saliva and oral cavity, e.g., ALDH2-deficient Asians, Caucasians with high activity ADH, smokers and heavy drinkers. Since L-cysteine is safe and effective in the concentrations used in our study, our finding opens a possibility for forthcoming clinical trials on cancer prevention.