Beer flavour instability: Unravelling formation and/or release of staling aldehydes

The sensory perception of food and beverages is expected to be pleasant, characteristic, and consistent at the moment of consumption. However, in particular, beer has a limited shelf life because of flavour deterioration during transport and storage. All beers are subject to so-called flavour instability: the instability of beer in terms of odour, aroma, taste and mouthfeel. Beer flavour instability is very important from an economic point of view as the brewing industry represents a gigantic market. Therefore, more specifically, a prolonged flavour stability is essential, as it may be critical for acceptance or rejection of the product by the consumer. Flavour instability of beer is characterized by physical and chemical changes over beer storage. Two phenomena are considered to be of utmost importance with respect to the ageing of beer: a decrease in the intensity and the quality of the typical beer bitterness and, an increase in the levels of ageing aldehydes (so-called beer staling aldehydes). This PhD is focused on the second issue, mainly the generation of staling aldehydes as a function of beer ageing. Aldehydes might be formed (in the beer bottle but equally well during the brewing process) de novo from precursors. In particular, three major routes for de novo formation of aldehydes have been proposed: I) fatty acid oxidation (resulting in e.g. trans-2-nonenal), II) Strecker degradation of amino acids (resulting in e.g. 2‑methylpropanal), and III) the Maillard reaction (resulting in e.g. furfural). Alternatively, the formation of non-volatile bound-state aldehydes during the malting and/or brewing process has been proposed as a potential pathway to explain increases in staling aldehydes during beer ageing. Indeed, pre-formed non-volatile bound‑state aldehydes may survive the brewing and fermentation processes, thereby ending up in the final beer and, via subsequent decomposition result in the release of aldehydes during beer ageing. Several previous studies suggested that bound-state aldehydes might be formed through reaction of aldehydes with bisulphite during fermentation or through reaction of aldehydes with amino acids, peptides, or proteins to form imines during malting and brewing. In addition, very recent work pointed to the bound form of aldehydes with cysteine, so-called cysteinylated aldehydes or 2-substituted 1,3-thiazolidine-4-carboxylic acids, as potentially responsible for the beer staling problem. The major objective of this PhD is to better understand the binding/release behaviour of the most relevant beer staling aldehydes and to further investigate the potential role of aldehyde adducts with cysteine and bisulphite in relation to beer ageing. The most important marker aldehydes to be considered in beer ageing are 2-methylpropanal, 2‑methylbutanal, 3-methylbutanal, methional, phenylacetaldehyde, furfural, hexanal, and trans‑2‑nonenal. Therefore, in the first stage of this PhD, from these particular aldehydes all adducts with cysteine and bisulphite, respectively, were synthesized. Structure verification of aldehyde adducts was obtained via NMR (1H-NMR and 13C-NMR) and high mass accuracy UHPLC-MS. Moreover, purity was assessed by high mass accuracy UHPLC-MS. Upon successful synthesis of all bound-state aldehydes, these compounds were used throughout the PhD as new reference material in order to be able to conduct all subsequent studies. To this end, first of all, a methodology to reliably quantify both free and bound-state aldehydes, based on HS-SPME-GC-MS and UHPLC-MS, respectively, was implemented. Validation comprised linearity, accuracy, limits of detection (LOD) and quantification (LOQ), and demonstrated that the methodologies to analyse both, volatiles and non‑volatiles, are appropriate for quantification purposes within the desired concentration range (µg/L) and, thus, represent a truly integrated analytical GC-MS/UHPLC-MS methodology. Upon synthesizing the most relevant adducts of beer staling aldehydes and establishing an optimized analytical methodology, these tools were then applied in studies on (1) model solutions and (2) real malting and brewery samples. In model solutions, the influence of pH and temperature on the behaviour of cysteinylated aldehydes was investigated. Free and cysteine-bound aldehydes were analysed via HS‑SPME‑GC‑MS and UPLC‑PDA, respectively. To monitor the behaviour of cysteinylated aldehydes under malting and brewery relevant conditions in model solutions over a time frame of 24 hours, pH values of 2.0, 4.4, 5.2, 6.0, and 9.0 and temperatures of 0 ºC, 20 ºC and 40 ºC were selected. Cysteinylated aldehydes showed degradation and concomitant release of corresponding free aldehydes at the pH values 4.4, 5.2 and 6.0, that are related to malting and brewing. Furthermore, stability of the adducts was observed at alkaline pH. A higher degradation rate was noticed at 40 ºC compared to 0 ºC. In summary, our study in model solutions points to the potential importance of cysteinylated aldehydes as a possible source of staling aldehydes during beer storage. Following the work in model solutions, the raw material malt, intermediate brewery samples, final beer and forced aged beer were analysed for both free and bound-state aldehydes by applying the integrated methodology developed previously. Free aldehydes were quantified in all tested samples (from malt to beer). Cysteinylated aldehydes were detected from malt until the start or end of wort boiling, depending on the nature of the aldehyde. However, bisulphite‑bound aldehydes were not found in any of the samples. Malt contained all free and cysteinylated aldehydes under study. Analysis of samples taken at the onset of the mash compared to malt samples, showed the highest levels of cysteinylated aldehydes at mashing‑in and highest levels of free aldehydes in malt, pointing to additional formation (next to formation in malting) of cysteine‑aldehyde adducts at mashing-in. Furthermore, decreasing concentrations of both free and cysteine-bound aldehydes were observed throughout the brewing process. An exception was found however for the free aldehyde furfural which increased in levels when strong heat-load was applied, i.e. during wort boiling and wort clarification. In fresh beer, only traces of free aldehydes were detected. However, levels in free aldehydes increased during beer ageing, in particular for 2‑methylpropanal and furfural. In contrast, cysteinylated aldehydes were not quantifiable in fresh beer samples, neither during beer ageing, with the exception of the cysteinylated form of 2‑methylpropanal. Although these findings do not designate cysteine‑ or bisulphite‑bound aldehydes as being directly responsible for beer flavour deterioration in the beer matrix, they add evidence to the concept of bound‑state aldehydes, and therefore to further consider and investigate the potential contribution of bound-state aldehydes in relation to flavour instability of food and beverages. Consequently, in the final experimental part of the PhD, to further investigate potential formation of bound-state aldehydes in the beer matrix, labelled aldehydes, i.e. 2‑methylbutanal-d3 and furfural-d3, were added in fresh commercial pale lager beer. Binding of the selected deuterated aldehydes could however not be demonstrated. Moreover, release of deuterated aldehydes upon beer ageing was also not detected. Lastly, an initial study of de novo formation of aldehydes from amino acids was performed in the beer matrix. To this end, the labelled amino acids valine-d8 and leucine-d3 were added to a fresh commercial pale lager beer and potential formation of deuterated aldehydes was monitored during beer ageing up to 3 months at 30 ºC. Increasing concentrations of both corresponding deuterated free aldehydes (2-methylpropanal and 3-methylbutanal) were found over ageing. This finding represents the first unambiguous evidence of de novo formation of aldehydes from amino acids in the beer matrix.status: accepte