Transient acetaldehyde production by SO2 producing S. cerevisiae promotes survival of Oenococcus oeni during co-fermentation

<p>Stuck or sluggish malolactic fermentation (MLF) can be problematic in limiting wine conditions, particularly white and sparkling base musts / wines. In these cases, knowledge of yeast-bacteria strain compatibility and the amount of sulfur dioxide (SO₂) a yeast strain produces are important considerations for successful MLF.</p> <p>Here, the effects of yeast-derived SO₂ production on <em>O. oeni</em> survival was investigated in laboratory- and pilot-scale co-fermentations in Chardonnay. Further to the <em>S</em>. <em>cerevisiae</em> strain affecting <em>O. oeni</em> survival and MLF, we show that SO₂ production by yeast (to approximately 65 mg/L) can be uncoupled from <em>O. oeni </em>survival in early stages of co-fermentation. Bacterial survival with certain SO₂-producing yeast strains was correlated with early, transient formation of a high concentration of acetaldehyde.  Upon co-inoculation, an extremely low concentration range (approximately 1–44 µg/L) of calculated molecular SO₂ is indicated to regulate <em>O</em>. <em>oeni</em> survival. <span>Possible strain-dependent sensitivity of <em>O. oeni</em> to bound SO₂ may also occur, although the extent and nature of such inhibition by the SO₂ adduct itself during co-fermentation remains unclear. </span></p> <p><span>Choice of co-inoculation yeast strain also influenced wine diacetyl concentration, </span>which was only detected in wines co-inoculated with high-SO₂-producing <em>S. cerevisiae </em>strains. <span>These wines also had comparatively high citation frequency for a buttery sensory attribute. </span>Both the SO₂ and acetaldehyde production capacity of yeasts are therefore seen as meaningful co-inoculation selection criteria. <span>The range of yeast strains suitable for MLF induction by co-inoculation could be widened to include SO₂-producing strains which transiently produce an early, high concentration of acetaldehyde. Reliance on total / bound SO₂ concentration may also potentially provide an inaccurate measure of SO₂ toxicity towards <em>O. oeni</em>, particularly for co-inoculation, and the impact of low, equilibrium concentrations of molecular SO₂ should also be considered. </span></p>