Influence of Cardiolipin on Lager Beer Dimethyl Sulfide Levels: A Possible Role Involving Mitochondria?

Abstract

Dimethyl sulfide (DMS) is an important flavor compound in American and European lagers, so brewers strive to control its level in their products. Although the main approach to controlling DMS involves judicious selection of malting and wort production regimes to control one of its precursors, S-methyl methionine (SMM), DMS formation during fermentation by yeast caused by an enzyme coined “dimethyl sulfoxide (DMSO) reductase” cannot be overlooked. During the process of altering the wort carbohydrate spectrum from one predominantly made up of maltose to one much richer in glucose through adjunct changes, a significant reduction in beer DMS levels was noted. An investigation into malt SMM levels demonstrated no significant reduction in this main precursor, suggesting other mechanisms could be involved. Since it is known that glucose inhibits the development of both mitochondria and their unique lipid, cardiolipin, we speculated that this organelle could be involved. Through a series of replicated incubation experiments aimed at altering cardiolipin content in yeast, it was found that beer DMS levels at the end of fermentation were moderately higher (approx. 10 μg/L) under incubation conditions that promote cardiolipin formation, i.e., incubation in glycerol and l-thyroxine. In comparison, conditions that limit cardiolipin formation (control and inositol incubation) resulted in lower DMS levels (P < 0.05). Assuming equal losses due to CO2 stripping during fermentation, these results suggest a possible role involving yeast mitochondria. Two suggestions are provided, of which we believe the latter is most likely based on DMSO- and methionine sulfoxide (MetSO)-addition experiments conducted within this study. First, if a true bacterial-like DMSO reductase protein were responsible, then defects in iron-sulfur cluster (ISC) formation or protein maturation would be likely, because ISCs are catalytic sites for electron transfer of Fe-S proteins. However, due to complete MetSO inhibition of DMSO reduction to DMS in the yeast strain studied, it seems more plausible that this was the result of decreased MetSO reductase activity, specifically an MsrB protein, which is speculated to be localized to the mitochondria, and that somehow the cardiolipin composition of the mitochondrial membrane is involved.