Correlation: Between Autochthonous Microbial Diversity and Volatile Metabolites During the Fermentation of Nongxiang Daqu

Abstract

Daqu is an important saccharifying and fermenting agent. It provides various microorganisms and enzymes for the fermentation of Baijiu and plays a vital role in the formation of Baijiu flavor. However, it is difficult to obtain information on microbial growth and metabolism in time for Daqu production. Therefore, the "Qu Xiang" obtained by smelling is an important index in the traditional production process to evaluate the microbial fermentation in the process of Daqu-making, "Qu Xiang" mainly represents the volatile flavor compounds in Daqu. The microbial diversity and volatile metabolites on 0, 6, 16, and 29 days of the fermentation process were measured using high-throughput sequencing and gas chromatography-mass spectrometry. Significant differences were found in the composition of the microbial community. Pseudomonas, Weissella, Bacillus, and Pelomonas were the main bacterial genera. Alternaria, Rhizopus, and Pichia are the main fungal genera. A total of 32 differential volatile metabolites were detected in samples at four time points using differential metabolic analysis. The correspondence of prevailing microorganisms with differential metabolites distinguished by Spearman correlation and two-way orthogonal partial least square analysis show that Saccharopolyspora exhibited a significant connection for the 12 differential metabolites. A significant positive correlation was observed between Rhizomucor and 13 different metabolites. These findings further understanding of the metabolism of microorganisms in Daqu fermentation and also help to control the microorganisms in the Daqu-making process, to obtain more stable Baijiu products.

Keywords: correlation analysis; differential metabolites; microbial communities; nongxiang daqu; process of fermentation.

FIGURE 1

A schematic of the production process of NongXiang Daqu (A). A Schematic diagram of increasing room (B). Temperature curve of the fermentation process (C).

FIGURE 2

Distribution of bacteria (A) and the fungi (B) on phylum level during the fermentation process of Daqu. And showing the changes of major bacteria (C) and fungi (D) in the samples during the fermentation on genus level.

FIGURE 3

Clustering analysis of bacteria (A) and the fungi (B) in different fermented samples (days 0, 6, 16 and 29)using principal component analysis. The differential bacteria (C) and fungi (D) in each time point of Daqu by using LEfSe analysis (LDA > 2, p < 0.05) (Four groups were set according to four time points).

FIGURE 4

Potential function prediction (based on the level-3 pathway in metabolite of KEGG database).

FIGURE 5

Kinetic comparison of volatile metabolites during fermentation. (A) Heatmap showing the Changes of different types of compounds in the fermentation process. (B) Clustering analysis of the volatile compounds in different fermented samples by orthogonal partial least squares analysis. (C) Distribution of different metabolites and value of variable importance in metabolite projection.

FIGURE 6

Heatmaps of Spearman correlations between dominant (A) bacterial and (B) fungal genera and differential metabolites during the fermentation process. Levels of significance are shown as follows: *0.01 < p ≤ 0.05; **0.001 < p ≤ 0.01,***p ≤ 0.001. The co-occurrence network diagram of a highly significant correlation between differential metabolites and differential bacteria (C) and fungi (D).

FIGURE 7

KEGG pathway enrichment of identified differential metabolites.

FIGURE 8

Schematic presentation of pathways, metabolites, enzymes, and genus related to the synthesis of Ethanol, Phenylethanol, Phenylacetaldehyde, and Lactic acids, ect in Daqu samples. The pathways were organized based on the KEGG (Kyoto Encyclopedia of Genes and Genomes).