HS-SPME-GC × GC/MS combined with multivariate statistics analysis to investigate the flavor formation mechanism of tank-fermented broad bean paste

Abstract

With the advancement of industrialization, tank fermentation technology is promising for Pixian broad bean paste. This study identified and analyzed the general physicochemical factors and volatile metabolites of fermented broad beans in a thermostatic fermenter. Headspace solid-phase microextraction (HS-SPME)-two-dimensional gas chromatography-mass spectrometry (GC × GC-MS) was applied to detect the volatile compounds in fermented broad beans, while metabolomics was used to explore their physicochemical characteristics and analyze the possible metabolic mechanism. A total of 184 different metabolites were detected, including 36 alcohols, 29 aldehydes, 26 esters, 21 ketones, 14 acids, 14 aromatic compounds, ten heterocycles, nine phenols, nine organonitrogen compounds, seven hydrocarbons, two ethers, and seven other types, which were annotated to various branch metabolic pathways of carbohydrate and amino acid metabolism. This study provides references for subsequent functional microorganism mining to improve the quality of the tank-fermented broad beans and upgrade the Pixian broad bean paste industry.

Keywords: Broad bean paste; GC x GC/MS; Metabolomics; Volatile compounds.

Fig. 1

The broad bean fermentation process.

Fig. 2

The physicochemical parameters for the constant temperature fermentation of broad beans. (A) Water content. (B) pH. (C) Total titratable acid. (D) Amino acid nitrogen. (E) Reducing sugars. The experiment for each parameter was performed in triplicate, while the standard error was visualized via an error bar.

Fig. 3

The results of the multivariate statistical analysis of the fermented broad bean samples. (A) The PCA analysis of the volatile compounds identified in the broad beans at different fermentation periods. (B) The OPLS-DA score plot. (C) The result of 200 times permutation testing for OPLS-DA. Five parallel experiments were performed for each sample.

Fig. 4

The proportion of volatile differential compounds.

Fig. 5

The heat map of the differential metabolite peak areas. A, B, C, D, E, and F represent the samples collected at 0 d, 6 d, 13 d, 20 d, 28 d, and 39 d of fermentation, respectively, while the numbers 1, 2, 3, 4, and 5 represent parallel experiments. The blue dots indicate the key flavor substances identified in previous studies. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Fig. 6

KEGG pathway-based differential metabolite analysis. (A) The KEGG pathway annotated in the yeast database. (B) The KEGG pathway annotated in the Bacillus database. (C) The KEGG pathway annotated in the Staphylococcus database. The importance of pathways was visualized by the size and color shade of the bubbles. (D) A comparison between the color of the fermented broad beans. Images of the pre-, mid-, and post-fermentation stages are shown from left to right. (E) The degradation of toluene.