Changes in the Human Metabolome Associated With Alcohol Use

A Review

Taija Voutilainen; Olli Kärkkäinen


Alcohol Alcohol. 2019;54(3):225-234. 

In This Article

Abstract and Introduction


Aims: The metabolome refers to the functional status of the cell, organ or the whole body. Metabolomic methods measure the metabolome (metabolite profile) which can be used to examine disease progression and treatment responses. Here, our aim was to review metabolomics studies examining effects of alcohol use in humans.

Methods: We performed a literature search using PubMed and Web of Science for reports on changes in the human metabolite profile associated with alcohol use; we found a total of 23 articles published before end of 2018.

Results: Most studies had investigated plasma, serum or urine samples; only four studies had examined other sample types (liver, faeces and broncho-alveolar lavage fluid). Levels of 51 metabolites were altered in two or more of the reviewed studies. Alcohol use was associated with changes in the levels of lipids and amino acids. In general, levels of fatty acids, phosphatidylcholine diacyls and steroid metabolites tended to increase, whereas those of phosphatidylcholine acyl-alkyls and hydroxysphingomyelins declined. Common alterations in circulatory levels of amino acids included decreased levels of glutamine, and increased levels of tyrosine and alanine.

Conclusions: More studies, especially with a longitudinal study design, or using more varied sample materials (e.g. organs or saliva), are needed to clarify alcohol-induced diseases and alterations at a target organ level. Hopefully, this will lead to the discovery of new treatments, improved recognition of individuals at high risk and identification of those subjects who would benefit most from certain treatments.


Metabolomics (metabonomics, metabolic profiling) is an omics approach that makes it possible to study the metabolome, i.e. metabolic changes in the body by measuring small molecules (metabolites) (Bujak et al., 2015). Unlike the genome and proteome, the metabolome directly represents the functional changes in cellular metabolism (Figure 1). Thus, it provides a view about the current physiological state of the sample. In the human circulation and organs, there are also metabolites produced by micro-organisms and metabolites originating directly from nutrition or other exogenous sources. Metabolomics has been widely utilized in investigating both physiological conditions and pathological states (Bujak et al., 2015). Consequently, metabolomics can also be used in biomarker research not only to identify early predictive markers and novel targets for therapies, but also to monitor disease progression and treatment outcomes. Furthermore, many metabolites have shown some potential to be used to modify phenotype, for example, to prevent organ damage or disease progression (Guijas et al., 2018).

Figure 1.

Metabolomic centric view of different omics-methods. Flow of information from genotype to phenotype is shown. Metabolomics is aiming to measure the metabolome, i.e. all of the small molecules (metabolites) produced by endogenous processes in a sample. Metabolomics methods also measure metabolites from different exposures, like nutrition or alcohol consumption, as well as the metabolites produced by microbiota. With respect to the omics, metabolomics focuses on functional changes and is closest to the phenotype. Metabolites can also affect epigenetic modulation, gene expression, and protein function (as can exogenous and microbiota produced compounds). The metabolome can be used to follow disease progression or responses to treatment, because it is altered by changes in endogenous processes and exogenous exposures.

The basic methods used in metabolomics research are nuclear magnetic resonance (NMR) and mass spectrometry (MS) coupled with either liquid or gas chromatography (LC or GC, respectively) (Ulaszewska et al., 2019). MS based methods are usually more sensitive than NMR based methods. In other words, MS can usually detect more metabolites than NMR. For example, from serum samples, NMR-based methods can measure tens to hundreds of compounds, whereas MS-based techniques are able to measure hundreds to thousands of metabolites. However, some important molecules, like lipoproteins, can be measured with NMR, but not with MS metabolomic methods. Therefore, the combination of both NMR and MS methods provides the widest possible coverage of different metabolites. Although the chromatographic and MS techniques are highly versatile, enabling measurement of different types of compounds, these approaches are subject to challenges in terms of reproducibility and variability between labs. NMR has lower costs per sample and is better suited for screening large numbers of samples. Furthermore, NMR metabolomic methods are usually quantitative. In contrast, MS-based methods can be either (a) untargeted and semi-quantitative (where the goal is to measure as many as possible metabolites) or (b) targeted and quantitative (where certain selected groups of metabolites are measured against standards). For a more detailed description and hints for study design see Ulaszewska et al. (2019).

Alcohol use disorder is a global health problem accounting for substantial losses for both individuals and societies (Griswold et al., 2018). The biochemical processes underpinning the adverse health effects are not yet fully understood. In this respect, metabolomics represents a novel approach to increasing our understanding of alcohol-related problems. Metabolomic methods enable follow-up of disease progression or an evaluation of how some treatment has affected the metabolome (Bujak et al., 2015). This increased understanding of pathology could lead to the discovery of possible new targets for therapies to prevent or mitigate alcohol caused health problems.

In addition, metabolomics could help in the search for identifying a biomarker profile or pattern; this could be used not only to predict disease progression or the response to medication (Nam et al., 2015; Hinton et al., 2017), but possibly also to quantify alcohol use. Unfortunately, many currently used biomarkers for alcohol use lack specificity and sensitivity. For example, the biomarkers measuring liver enzymes provide information only about the functionality of the liver but not the cause of damage. Another biomarker, the percentage of carbohydrate deficient transferrin (%CDT), can be elevated due to severe liver disease or due to pregnancy (Gough et al., 2015; Bortolotti et al., 2018). Recently, phosphatidylethanols have been introduced as selective biomarkers for alcohol use, but they need to be measured from samples not collected routinely (whole blood or tissues) because phosphatidylethanols are an integral part of cell membranes (Zheng et al., 2011). Clearly, there is a need to devise biomarkers which would indicate the quantity of alcohol use, and that these could be measured routinely from serum or plasma. This is especially true with sensitive populations like pregnant women.

The aim of this review was to examine and compare the current publications about how alcohol use influences metabolome in humans. Furthermore, we will highlight some future directions for this field.