Selection and characterization of novel malolactic starters from Rioja Alavesa region: in the pursuit of free-histamine wine

Abstract

Biogenic amines (BAs) are considered a high priority issue for wine industry in terms of product quality and safety. They may cause several adverse symptoms which are enhance in susceptible individuals. The inoculation of safe malolactic starters is one possible strategy to prevent the accumulation of these compounds in wine. In this sense, the main objective of this work was to identify and characterize autochthonous lactic acid bacteria (LAB) strains from Rioja Alavesa region lacking the ability to produce BAs and owning desired technological and sensorial properties. Firstly, after the analysis of 70 red wines it was obtained a snapshot of the current situation of BAs levels in Rioja Alavesa region. The most abundant was putrescine (14,85 ± 8,9 mg/L) followed by histamine (4,43 ± 2,8 mg/L), tyramine (3,29 ± 3,28 mg/L) and cadaverine (2,14 ± 1,58 mg/L). Considering wine sectors demand to reduce BAs levels to minimum, the pursue of novel malolactic starters was initiated. From near 300 isolates, 27 genotypes belonging to Pediococcus parvulus (3), Lactobacillus plantarum (1), Lactobacillus mali (3), Lactobacillus hilgardii (3) and Oenococcus oeni (17) species were identified. In this regard, it was confirmed the great predominance of O. oeni. Among LAB species, only L. hilgardii strains were able to produce putrescine via the agmatine deiminase pathway. The technological characterization of LAB strains elucidated the great suitability of O. oeni species against typical harsh conditions found in wine as well as their better performance at conducting the malolactic fermentation (MLF) over the rest of species. Further characterization over O. oeni strains elucidated their ability to retain different glycosidase (¿-glucosidase, ß-glucosidase, ß-xylosidase and ¿-arabinosidase) and esterase activities under winemaking conditions. The prospective use of O. oeni strains as malolactic starters was first examined at laboratory scale microvinifications. No production of BAs was detected, and the evolution of aromatic compounds showed that main changes after MLF occurred for both ethyl and acetate esters. Principal component analysis classified the strains in two distinct groups, highlighting the specific impact of each strain on wine aroma profile. Most promising five O. oeni strains were submitted to co-inoculation and sequential inoculation fermentation processes. Co-inoculation led to the prompt consecution of winemaking process and no production of BAs was detected during any MLF. In co-inoculated wines the perception of ripe fruit aroma was extensively perceived in comparison with their respective sequentially inoculated wines. Finally, it was elucidated the suitability of the strain P2A to work in large scale fermentations at winery, resulting an advantageous alternative to significantly reduced the overall winemaking time as well as to better control the fermentative process