associations with maltose-negative yeasts such as Candida humilis (De
Vuyst and Vancanneyt, 2007; Vigentini et al., 2014).
Leavening trials performed using the unconventional combination
between Lactobacillus and Zymomonas always showed CO2 production
levels lower than S. cerevisiae alone. Nevertheless, the observed performance should more properly resemble the behavior of a typical sourdough not containing S. cerevisiae (Martinez-Anaya et al., 1990).
All the tested microbial associations gave higher CO2 production
levels than the mathematical sum of the CO2 produced in trials carried
out by the single bacteria. The calculated productivity gain showed positive performance in the first 180–240 min leavening; the subsequent
efficiency loss may be due to several factors, above all glucose deficiency
for Z. mobilis, as well as decrease of dough pH that can negatively affect
both Lactobacillus and Zymomonas metabolism.
Actually, Z. mobilis pH growth range is 4.5–6.5, with about 40% of
population resistant at pH 3.5 (Panesar et al., 2006; Sahm et al., 2006);
L. sanfranciscensis pHopt is around 5.5, with a pHmin 3.6 (Gänzle et al.,
1998).
As regards ethanol productivity (matching CO2 production) literature
data report the maximum yield when pH is in the range 5.0–5.5 (Khoja
et al., 2015) and very low yield at pH 4.0 (Gunasekaran et al., 1986).
In our leavening trials, when L. sanfranciscensis is present at higher
concentration (i.e. Lb:Zym 1:1 and 10:1) the productivity gain is evident
in the first 180 min, when glucose is efficiently released in dough and
consumed by Zymomonas for alcoholic fermentation (maltose and glucose time course, Fig. 4). When incubation proceeds, CO2 productivity
decreased to values lower than the theoretical ones (Table 2), dough
pH reached values lower than 4, and Z. mobilis population significantly
decreased (Table 3 and Table 4); these data together with the fact that