Investigation of blueberry juice fermentation by mixed probiotic strains: Regression modeling, machine learning optimization and comparison with fermentation by single strain in the phenolic and volatile profiles

Highlights

• Three probiotics were combined to ferment blueberry juice.

• Regrssion modeling and genetic algorithm were used to optimize inoculation proportion.

• The combined L. plantarum and L. fermentum at 0.5:0.5 enriched phenolics best.

• Mixed fermented juices had highest rutin and quercetin-3-rhamnoside contents.

• Mixed fermentation produced more alcohols and ketones than L. fermentum fermentation.

Abstract

Three strains, including L. fermentum, L. plantarum and S. thermophilus, were combined to ferment blueberry juice. Through the sequential simplex lattice mixture design, regression modeling and genetic algorithm optimization, it was found that the combination of S. thermophilus with either L. fermentum or L. plantarum weakened the capacity of Lactobacillus strains to enrich phenolics, and the combinations of these strains had no synergistic effect of synthesizing lactic acid. The resulting optimal inoculation proportion to enrich phenolics was the mixed L. fermentum and L. plantarum at 0.5:0.5. After fermentation for 48 h, total phenolic, ferulic acid, rutin, and quercetin-3-rhamnoside of mixed fermented samples were 82.19 %, 15.22 %, 79.08 % and 98.59 % higher than the unfermented juice, and their contents were all highest among the fermented samples. Moreover, the samples fermented by mixed strains possessed higher amounts of 3-methyl-1-butanol, 2-methyl-1-propanol, 2-heptanone and 2-pentanone than samples fermented by L. fermentum, S. thermophilus and unfermented samples.

Introduction

Probiotic fermented fruit and vegetable (F&V) juice, recognized as a nutritive product, is gaining more public concern and acceptability. Probiotics are defined as “Live microorganisms which when administered in adequate amounts confer a health benefit on the host” (FAO, 2006). The lactic acid bacterium is an important member of probiotics that have already been widely applied in the food industry. Probiotic fermentation has a profound influence on food composition. Briefly, proteases of lactic acid bacteria can decompose the corresponding substrates into free amino acids, improving the freshness of fermented products (Ren, & Li, 2022). Also, lactic acid bacteria can produce alcohols, esters, and other aromatic components to intensify the mellow fruit aroma (Li et al., 2018). Moreover, the complicated derivation reactions among polyphenols occur in the presence of lactic acid bacteria, improving the F&V juice taste and nutritional value (Sevindik et al., 2022). Many other benefits also exist with the probiotic fermentation of F&V juices, not limiting to the list aspects.

Lactic acid bacteria are widely distributed in the environments, and the microbial properties vary with the host materials. The typical food-grade lactic acid bacteria include Lactobacillus plantarum (Liu, Shao, Zhang, & Kang, 2020), Lactobacillus fermentum, Streptococcus thermophilus and others (Szutowska, 2020). In the microbial community, among different microbial species, there might be interactions if the microorganisms grow in the same environment, i.e. symbiosis and antagonism. Although most researches on mixed fermentation of F&V juices focus on fruit wine, increasing evidence reveals that the mixed fermentation of Saccharomyces cerevisiae, non-Saccharomyces yeasts and non-yeast strains can improve the wine quality substantially. For example, the volatile flavoring substances such as ethyl ester and fatty acids in white wine were enriched by the mixed fermentation of Pichia fermentans and S. cerevisiae (Ma, Yan, Wang, Zhang, & Tao, 2017). Also, the contents of anthocyanins and tannins in red wine were increased by the simultaneous fermentation of P. fermentans and S. cerevisiae (Kong et al., 2019). Besides, the mixed fermentation of bergamot juice by three types of Lactobacillus plantarum possessed stronger antioxidant capacity than the juices fermented by each species separately (Hashemi & Jafarpour, 2020). However, detailed studies about the F&V juice fermentation by various microbial strains are still scarce and need to be explored.

On the other hand, many F&V juices feature a low pH and high amount of soluble fiber and bioactive substances, i.e. polyphenols. This environment is adverse to the growth and metabolism of lactic acid bacteria. Several autochthonous L. fermentum and L. plantarum were isolated from different F&V in our previous work, some of which were potent in the bio-transformation of polyphenols in blueberry juices (Li, Tao et al., 2021). Meanwhile, in another study, we found that the fermentation of S. thermophilus improved the nutritional and sensory quality of blackberry and blueberry juices (Wu et al., 2021). Thus, it is important to examine the properties of the combination of these lactic acid bacterial species for the fermentation of F&V juices.

In recent years, mathematical tools, particularly experimental design and marchine learning algorithms, have been continuously applied in food science research. For instance, the combined mixture design (i.e. simplex lattice mixture design, simplex centroid design) and polynomial regression have been utilized to study the problems of food formulas (Rad, Pirouzian, Toker, & Konar, 2019). Meanwhile, the model optimization by genetic algorithm (GA) has been employed to accurately find the best solution to food problems, such as extraction, formula establishment, etc. (Tao et al., 2017). These methodologies can be considered when studying blueberry fermentation by mixed strains.

In this investigation, three previously studied probiotics (L. fermentum, L. plantarum and S. thermophilus) were selected as mixed fermentation cultures to ferment blueberry juice. The influence of mixed fermentation on the lactic acid accumulation and polyphenols enrichment were explored and optimized by the sequential simplex lattice mixture design, polynomial regression and GA optimization. Then, metabonomic data about phenolics, organic acids, and volatiles from mixed fermentation were compared with individual fermentation.