Microbial Community Dynamics and the Correlation between Specific Bacterial Strains and Higher Alcohols Production in Tartary Buckwheat Huangjiu Fermentation

Tartary buckwheat is a healthy grain rich in nutrients and medicinal ingredients and consequently is commonly used for Huangjiu brewing. In order to reveal the correlation between microbial succession and higher alcohols production, in this study, Huangjiu fermentation was conducted using Tartary buckwheat as the raw material and wheat Qu as the starter culture. Microbial community dynamics analysis indicated that the bacterial diversity initially decreased rapidly to a lower level and then increased and maintained at a higher level during fermentation. Lactococcus was the dominant bacteria and Ralstonia, Acinetobacter, Cyanobacteria, and Oxalobacteraceae were the bacterial genera with higher abundances. In sharp contrast, only 13 fungal genera were detected during fermentation, and Saccharomyces showed the dominant abundance. Moreover, 18 higher alcohol compounds were detected by GC-MS during fermentation. Four compounds (2-phenylethanol, isopentanol, 1-hexadecanol, and 2-phenoxyethanol) were stably detected with high concentrations during fermentation. The compound 2-ethyl-2-methyl-tridecanol was detected to be of the highest concentration in the later period of fermentation. Correlation analysis revealed that the generation of 2-phenylethanol, isopentanol, 1-hexadecanol, and 2-phenoxyethanol were positively correlated with Granulicatella and Pelomonas, Bacteroides, Pseudonocardia and Pedomicrobium, and Corynebacterium, respectively. The verification fermentation experiments indicated that the improved wheat Qu QT3 and QT4 inoculated with Granulicatella T3 and Acidothermus T4 led to significant increases in the contents of 2-phenylethanol and pentanol, as well as isobutanol and isopentanol, respectively, in the Tartary buckwheat Huangjiu. The findings benefit understanding of higher alcohols production and flavor formation mechanisms in Huangjiu fermentation.


Introduction
Huangjiu (Chinese rice wine) is one of the most famous traditional fermented liquors in China and has a brewing history of thousands of years [1]. Generally, Huangjiu brewing uses various grains (rice or millet) as raw materials and the wheat Qu (solid compounds consisting of wheat, various microorganisms, and complex enzyme systems) as the fermentation starter culture, which contribute to the unique organoleptic qualities and characteristics [2]. Huangjiu is popular due to not only its desirable flavor but also rich nutrition and health-promoting ingredients, including polyphenols, polysaccharides, oligosaccharides, peptides, organic acids, and trace elements [3]. Tartary buckwheat is regarded as a healthy grain rich in nutrients and medicinal ingredients and could be of benefit in the prevention of cardiovascular and cerebrovascular diseases [4]. Hence, Tartary buckwheat is commonly used as the raw material for healthcare Huangjiu brewing.

Tartary Buckwheat Huangjiu Fermentation and Samples Preparation
One kilogram of Tartary buckwheat (Zhangjiakou, Hebei, China) was washed and soaked in clean water for 24 h and then cooked in clean water (three times the volume) for 1 h. Glucoamylase (0.15%, w/w) was added to the well-cooked Tartary buckwheat for saccharification for 60 min. The wheat Qu (15%, w/w, BeiZong Huangjiu Company, Zhangjiakou, Hebei, China) and yeast (0.15%, w/w) were added to the saccharified mixture and mixed thoroughly for fermentation. Primary fermentation was performed in a tank at 30 • C for 10 days, and secondary fermentation continued at 15 • C for another 30 days. The fermentation samples were prepared every two days for higher alcohols determination and microbial DNA extraction. Metagenomic DNA was extracted from the fermentation samples using E.Z.N.A. soil DNA kit (Omega Bio-tek, Norcross, GA, USA). DNA quantification was conducted using NanoDrop 2000 Spectrophotometer (Thermo Scientific, Wilmington, NC, USA). The 16S rDNA and ITS1 amplicons were generated by PCR from 50 ng of genomic DNA in GeneAmp 9700 System (ABI, Los Angeles, CA, USA), respectively. The V3-V4 hypervariable regions of bacterial 16S rDNA were amplified using the forward primer F338 (5 -ACTCCTACGGGAGGCAGCAG-3 ) and the reverse primer R806 (5 -GGACTACHVGGGT WTCTAAT-3 ) [19]. The PCR reaction was performed with the following cycling conditions: 95 • C for 5 min; 25 cycles of 95 • C for 30 s, 55 • C for 30 s, and 72 • C for 40 s; and 72 • C for 10 min. In addition, the hypervariable ITS region of relatively conserved fungi rRNA was amplified using the forward primer ITSF (5 -CTTGGTCATTTAGAGGAAGT-3 ) and the reverse primer ITSR (5 -GCTGCGTTCTTCATCGATGC-3 ) [20]. The PCR reaction was performed with the following cycling conditions: 95 • C for 5 min; 30 cycles of 95 • C for 30 s, 61 • C for 30 s, and 72 • C for 45 s; and 72 • C for 10 min. The purified amplicons in equal molar ratios were loaded on Illumina MiSeq platform (Illumina, San Diego, CA, USA). A 2 × 300 paired-end configuration was conducted for sequencing. Image analysis and base calling were performed using the MiSeq Control Software 3.1.

Sequencing Data Analysis
The sequences were filtered by Trimmomatic for analysis. The effective sequences preclustered at a 97% identity were classified into operational taxonomic units (OTUs) by Usearch (http://drive5.com/uparse/). The OTUs of 16S rRNA were assigned taxonomic categories using the ribosomal database program (RDP) classifier at a confidence threshold of 70% and then predicted to the species level using the Silva 16S rRNA database. The OTUs of ITS were analyzed using the Unite program (http://unite.ut.ee/index.php). The calculations of the alpha and beta diversity statistics were conducted after sequence rarefication. Microbial community diversity analysis was performed by the alpha diversity using the Simpson index. Species correlation was calculated using the Pearson correlation coefficient.

Higher Alcohols Determination during Tartary Buckwheat Huangjiu Fermentation
Higher alcohols produced during the Tartary buckwheat Huangjiu fermentation were determined by headspace solid phase microextraction (HS-SPME) sampling combined with gas chromatography and mass spectrometry (GC-MS) analysis. The fermentation sample was centrifuged at 8000× g at 4 • C for 10 min, and the supernatant was collected and filtered through a 0.22 µm filter. The filtered supernatant (0.5 mL), 15% ethanol (5.44 mL), and the internal standard substance 2-octanol (8.8 mg/L, 60 µL) were added in a 15 mL headspace vial. Analytes were extracted by SPME for 45 min in water bath (50 • C) with ultrasonication using an adapter equipped with a 50/30 µm DVB/CAR/PDMS fiber (1 cm, Supelco Inc., Bellefonte, PA, USA). The fiber was then inserted into the injection port of the GC for thermal desorption at 250 • C for 7 min.
GC-MS analysis was conducted on Shimadzu GC-MS 2010Plus system (Shimadzu, Kyoto, Japan) with a DB column (30 m × 0.25 mm i.d., 0.25 µm film thickness). Helium (99.999 purity) was used as the column carrier gas at a constant flow rate of 1.0 mL/min. The injector and detector temperature was set at 250 • C. The oven temperature was held at 40 • C for 3 min, raised to 100 • C at a rate of 6 • C/min, and then raised to 230 • C at a rate of 10 • C/min and was held at 230 • C for 7 min. The MS transfer line temperature and ion source temperature were set at 230 • C. Electron ionization (EI, 50 µA electric current, 70 eV electron) was used as the ionization mode for GC-MS analysis. The ion currents within the mass range from 33 to 400 were monitored for analysis. The content of higher alcohols was calculated by the internal standard substance 2-octanol detected by GC-MS according to the following formula.
Ac Ais Cis where C is the content of higher alcohol, µg/L; Cis is the content of 2-octanol, µg/L; Ac is the peak area of higher alcohol; and Ais is the peak area of 2-octanol.

Correlation Analysis between Microbes and High Alcohols Production during Tartary Buckwheat Huangjiu Fermentation
Significant correlations between microbes at the genus level and higher alcohols production were established by the Pearson correlation coefficient. The Benjamini-Hochberg procedure was used to adjust p-values (p ≤ 0.05). The correlation network was built using R Programming.

Verification of the Correlation between the Wheat Qu Bacterial Strains and High Alcohols Production during Tartary Buckwheat Huangjiu Fermentation
In order to verify the correlation between specific microbial strains and high alcohols production, four bacterial strains (Lactococcus T1, Ralstonia T2, Granulicatella T3, and Acidothermus T4) isolated from the wheat Qu were used for the improvement of wheat Qu making. Briefly, the smashed wheat was mixed with clean water (20%, w/w) and the fresh cultures (1%, v/v) of different bacterial strains T1, T2, T3, and T4, respectively; the mixture was stirred thoroughly and compacted into piles, and spontaneous fermentation was performed at 40~45 • C for 30 days. The original wheat Qu (Q) and the improved wheat Qu added with bacterial strains (QT1, QT2, QT3, and QT4) were used for the Tartary buckwheat Huangjiu fermentation as previously described, respectively. Higher alcohols production during Huangjiu fermentation was determined, as previously described.

Dynamic Change in Bacterial Community during Tartary Buckwheat Huangjiu Fermentation
To reveal the dynamic change in the bacterial community during the Tartary buckwheat Huangjiu fermentation, the microbial DNA from different fermentation samples was subject to high-throughput sequencing and analysis. The data showed that 981 bacterial OTUs were classified from 1,002,498 qualified sequencing fragments in all fermentation samples. Sequence analysis indicated that those bacterial OTUs could be assigned to 30 phyla, 66 classes, 133 orders, 232 families, and 482 genera. As shown in Figure 1, Firmicutes and Proteobacteria had the highest abundance at the level of phylum during the Tartary buckwheat Huangjiu fermentation. Firmicutes dominated at the beginning of fermentation, where their percentage reached 85.52% on day two, and then they decreased gradually. Proteobacteria showed an opposite dynamic trend in comparison with Firmicutes. In addition, the abundance of Actinobacteria and Bacteroidetes was relatively high. At the genus level, Lactococcus was the dominant microbe, which rapidly increased to a peak percentage of 80% on day two and then decreased gradually during fermentation ( Figure 2). Ralstonia, Acinetobacter, Cyanobacteria, and Oxalobacteraceae also had a higher abundance. Species correlation analysis ( Figure 3) indicated that Lactococcus was negatively correlated with 20 bacterial genera, such as Lactobacillus and Sphingobium. There was a positive correlation among Escheria-shigella, Sphingomonas, Massilia, Lysobacter, Nocardioides, and Enterobacteriaceae. In the correlation network, Rhizobium, Ochrobactrum, and Streptomyces played an important role as the connection hub. Alpha diversity analysis showed that the Simpson index reached higher values on day two and four of fermentation, revealing that the bacterial diversity first rapidly decreased to a lower level and then increased and maintained at a higher level during fermentation ( Figure 4). The gene functions prediction related to KEGG pathways based on sequencing data analysis revealed that different samples exhibited similar gene functions during fermentation. But large differences were observed in the relative abundance of gene functions from different samples. As shown in Figure 5, the main metabolic pathways of bacteria during fermentation were mainly enriched in membrane transport, amino acid metabolism, and carbohydrate metabolism. The abundances in carbohydrate metabolism and amino acid metabolism gradually increased at first and then decreased at the end of fermentation in 10 days, which was in accordance with the dynamic change in higher alcohols generation. This revealed that bacteria utilized carbohydrates from Tartary buckwheat and proliferated quickly along with the fermentation and simultaneously produced higher alcohols from amino acid metabolism via the Ehrlich pathway.
positive correlation among Escheria-shigella, Sphingomonas, Massilia, Lysobacter, Nocardioides, and Enterobacteriaceae. In the correlation network, Rhizobium, Ochrobactrum, and Streptomyces played an important role as the connection hub. Alpha diversity analysis showed that the Simpson index reached higher values on day two and four of fermentation, revealing that the bacterial diversity first rapidly decreased to a lower level and then increased and maintained at a higher level during fermentation ( Figure 4). The gene functions prediction related to KEGG pathways based on sequencing data analysis revealed that different samples exhibited similar gene functions during fermentation. But large differences were observed in the relative abundance of gene functions from different samples. As shown in Figure 5, the main metabolic pathways of bacteria during fermentation were mainly enriched in membrane transport, amino acid metabolism, and carbohydrate metabolism. The abundances in carbohydrate metabolism and amino acid metabolism gradually increased at first and then decreased at the end of fermentation in 10 days, which was in accordance with the dynamic change in higher alcohols generation. This revealed that bacteria utilized carbohydrates from Tartary buckwheat and proliferated quickly along with the fermentation and simultaneously produced higher alcohols from amino acid metabolism via the Ehrlich pathway.  oides, and Enterobacteriaceae. In the correlation network, Rhizobium, Ochrobactrum, and Streptomyces played an important role as the connection hub. Alpha diversity analysis showed that the Simpson index reached higher values on day two and four of fermentation, revealing that the bacterial diversity first rapidly decreased to a lower level and then increased and maintained at a higher level during fermentation (Figure 4). The gene functions prediction related to KEGG pathways based on sequencing data analysis revealed that different samples exhibited similar gene functions during fermentation. But large differences were observed in the relative abundance of gene functions from different samples. As shown in Figure 5, the main metabolic pathways of bacteria during fermentation were mainly enriched in membrane transport, amino acid metabolism, and carbohydrate metabolism. The abundances in carbohydrate metabolism and amino acid metabolism gradually increased at first and then decreased at the end of fermentation in 10 days, which was in accordance with the dynamic change in higher alcohols generation. This revealed that bacteria utilized carbohydrates from Tartary buckwheat and proliferated quickly along with the fermentation and simultaneously produced higher alcohols from amino acid metabolism via the Ehrlich pathway.

Dynamic Change in Fungal Community during Tartary buckwheat Huangjiu Fermentation
In sharp contrast to the complex bacterial community, only 20 fungal OTUs were classified from 659,656 qualified sequencing fragments in all the fermentation samples. Those OTUs could be assigned to 1 kingdom, 4 phyla, 7 classes, 7 orders, 11 families, and 13 genera. At the phylum level, Ascomycota absolutely dominated in abundance in all the fermentation samples (Figure 6). At the genus level, Saccharomyces showed absolute dominance in abundance in all the fermentation samples (Figure 7), but Aspergillus and Malassezia with extremely low abundance were detected at the beginning of fermentation. Correlation analysis (Figure 8) indicated that Wickerhamomyces played an important role as

Dynamic Change in Fungal Community during Tartary Buckwheat Huangjiu Fermentation
In sharp contrast to the complex bacterial community, only 20 fungal OTUs were classified from 659,656 qualified sequencing fragments in all the fermentation samples. Those OTUs could be assigned to 1 kingdom, 4 phyla, 7 classes, 7 orders, 11 families, and 13 genera. At the phylum level, Ascomycota absolutely dominated in abundance in all the fermentation samples ( Figure 6). At the genus level, Saccharomyces showed absolute dominance in abundance in all the fermentation samples (Figure 7), but Aspergillus and Malassezia with extremely low abundance were detected at the beginning of fermentation. Correlation analysis (Figure 8) indicated that Wickerhamomyces played an important role as the connection hub in the correlation network, which was positively correlated with Penicillium and Humicola. Rhizomucor was found to be located near the center of the correlation network and might play a key role in maintaining the tight connection of the whole network. A mutual positive correlation existed among Rhizomucor, Thermoascus, and Aspergillus. Alpha diversity analysis showed that the Simpson index increased slightly during fermentation, revealing that the fungal diversity gradually decreased. However, no significant difference in the fungal diversity was observed in different fermentation samples (Figure 9). the connection hub in the correlation network, which was positively correlated with Penicillium and Humicola. Rhizomucor was found to be located near the center of the correlation network and might play a key role in maintaining the tight connection of the whole network. A mutual positive correlation existed among Rhizomucor, Thermoascus, and Aspergillus. Alpha diversity analysis showed that the Simpson index increased slightly during fermentation, revealing that the fungal diversity gradually decreased. However, no significant difference in the fungal diversity was observed in different fermentation samples (Figure 9).    the connection hub in the correlation network, which was positively correlated with Penicillium and Humicola. Rhizomucor was found to be located near the center of the correlation network and might play a key role in maintaining the tight connection of the whole network. A mutual positive correlation existed among Rhizomucor, Thermoascus, and Aspergillus. Alpha diversity analysis showed that the Simpson index increased slightly during fermentation, revealing that the fungal diversity gradually decreased. However, no significant difference in the fungal diversity was observed in different fermentation samples ( Figure 9).

Higher Alcohols Production during Tartary buckwheat Huangjiu Fermentation
Eighteen higher alcohol compounds were detected by GC-MS during the 10 days of the Tartary buckwheat Huangjiu fermentation, as shown in Table 1. Most higher alcohol compounds appeared in the early four days of fermentation, and then several compounds dominated till the end of fermentation. Generally, four compounds (2-phenylethanol, isopentanol, 1-hexadecanol, and 2-phenoxyethanol) were stably detected with high concentrations in samples of different fermentation stages, which might significantly affect the flavor of Huangjiu. In addition, abundant 2-ethyl-2-methyl-tridecanol was detected during the later period of fermentation, whose concentration was the highest among all the higher alcohols. Pentanol with a quite high concentration was only detected on the eighth day of fermentation.

Higher Alcohols Production during Tartary buckwheat Huangjiu Fermentation
Eighteen higher alcohol compounds were detected by GC-MS during the 10 days of the Tartary buckwheat Huangjiu fermentation, as shown in Table 1. Most higher alcohol compounds appeared in the early four days of fermentation, and then several compounds dominated till the end of fermentation. Generally, four compounds (2-phenylethanol, isopentanol, 1-hexadecanol, and 2-phenoxyethanol) were stably detected with high concentrations in samples of different fermentation stages, which might significantly affect the flavor of Huangjiu. In addition, abundant 2-ethyl-2-methyl-tridecanol was detected during the later period of fermentation, whose concentration was the highest among all the higher alcohols. Pentanol with a quite high concentration was only detected on the eighth day of fermentation.

Higher Alcohols Production during Tartary Buckwheat Huangjiu Fermentation
Eighteen higher alcohol compounds were detected by GC-MS during the 10 days of the Tartary buckwheat Huangjiu fermentation, as shown in Table 1. Most higher alcohol compounds appeared in the early four days of fermentation, and then several compounds dominated till the end of fermentation. Generally, four compounds (2-phenylethanol, isopentanol, 1-hexadecanol, and 2-phenoxyethanol) were stably detected with high concentrations in samples of different fermentation stages, which might significantly affect the flavor of Huangjiu. In addition, abundant 2-ethyl-2-methyl-tridecanol was detected during the later period of fermentation, whose concentration was the highest among all the higher alcohols. Pentanol with a quite high concentration was only detected on the eighth day of fermentation.

Correlation Analysis of Microbes and Higher Alcohols Production during Tartary Buckwheat Huangjiu Fermentation
The production of 18 higher alcohol compounds was correlated with 294 microbial genera (p ≤ 0.05). The generation of 2-phenylethanol was positively correlated with seven genera, including Granulicatella, Pelomonas, Coprococcus, Oscillibacter, Ruminiclostridium, Ruminococcaceae, and Deltaproteobacteria. The production of pentanol was positively correlated with 72 genera, such as Rhodopseudomonas, while isopentanol was positively correlated with 31 genera, such as Bacteroides. Though Clostridium is the natural producer of butanol, Clostridium sensu stricto and 51 other genera, such as Acidothermus, were found to be positively correlated with isobutanol generation, but a negative correlation was observed between Ignavibacterium and isobutanol production. The appearance of 2-phenoxyethanol was positively correlated with 79 genera, such as Corynebacterium. As for alkanol compounds, 1-hexadecanol was positively correlated with 72 genera, such as Pseudonocardia and Pedomicrobium, and 1-hendecanol was positively correlated with 72 genera, such as Agromyces.

Effects of Improved Wheat Qu Inoculated with Specific Bacterial Strains on Higher Alcohols Production in Tartary Buckwheat Huangjiu Fermentation
The aforementioned results indicated that Lactococcus and Ralstonia were the dominant bacterial genera during Huangjiu fermentation, and Granulicatella and Acidothermus were positively correlated with the production of 2-phenylethanol and isobutanol. In order to verify the roles of specific bacterial strains in higher alcohols production, the bacterial strains Lactococcus T1, Ralstonia T2, Granulicatella T3, and Acidothermus T4 isolated from the wheat Qu were used for the improved wheat Qu making and consequent Huangjiu fermentation. The fermentation results (Table 2) showed that the improved wheat Qu QT1 and QT2 inoculated with Lactococcus T1 and Ralstonia T2 hardly led to significant differences in higher alcohols production. In sharp contrast, the improved wheat Qu QT3 and QT4 inoculated with Granulicatella T3 and Acidothermus T4 indeed resulted in significant increases in the contents of several specific higher alcohol compounds and total higher alcohols in Huangjiu. The additional inoculation of Granulicatella T3 in QT3 contributed to the higher contents of 2-phenylethanol and pentanol, and Acidothermus T4 in QT4 gave rise to more isobutanol and isopentanol generation. These results verified the positive correlations between Granulicatella and 2-phenylethanol production and Acidothermus and isobutanol generation during the Tartary buckwheat Huangjiu fermentation. It also revealed that Granulicatella and Acidothermus might get involved in the biosynthesis of multiple higher alcohol compounds during Huangjiu fermentation.

Discussion
In different regions, Chinese Huangjiu has distinguish local flavor characteristics due to the diversities in raw materials, microbial starter cultures (Qu), and brewing techniques [21][22][23]. Therefore, understanding the correlation between flavor characteristics and raw materials and microbial metabolisms is indispensable for the quality control of Huangjiu products. In this study, the featured Huangjiu made from Tartary buckwheat rich in functional components was used for the analysis of microbial community dynamics and higher alcohols production to establish a correlation between flavor formation and raw materials and microbial fermentation.
The microbes in Qu are the core driving force for Huangjiu brewing and they play an essential role in flavor compounds generation [23][24][25]. Previous studies showed that significant variances in microbial communities were observed in different Huangjiu Qu samples, and the dominant microbes with high abundances included Bacillus, Weissella, Pediococcus, Pantoea, Lactobacillus, Rhizopus, and Saccharomyces [21,23,24]. However, the dominant microbes during the Tartary buckwheat Huangjiu fermentation were quite different, which included Lactococcus, Ralstonia, Acinetobacter, Cyanobacteria, Oxalobacteraceae, and Saccharomyces. The significant variety evidently reflected the distinctive features of the local wheat Qu and raw materials used for Huangjiu brewing.
The formation of flavor compounds in Huangjiu were closely related with the raw materials and starter cultures used in the fermentation [22,23]. During the Tartary buckwheat Huangjiu brewing, several higher alcohol compounds, including 2-phenylethanol, 2phenoxyethanol, 2-ethyl-2-methyl-tridecanol, and 1-hexadecanol, were constantly detected from the beginning to the end of the fermentation, and their concentrations remarkably increased along with the fermentation. It revealed that those compounds were generated from both Tartary buckwheat and starter cultures fermentation. The compounds that only appeared on day 0 probably were released from Tartary buckwheat. In addition, some compounds arose along with fermentation and disappeared afterwards, which probably were the intermediate metabolites produced by fermentation microbes. The specific contribution of Tartary buckwheat to the flavor formation of the fermented Huangjiu remains a mystery due to the lengthy fermentation process driven by the wheat Qu and various microbes, as well as the complex chemical components of Tartary buckwheat. Tartary buckwheat contains proteins, polysaccharide, polyphenolic compounds, multiple vitamins, cellulose, weakly alkaline starch, various trace elements, dietary fiber, and various kinds of essential amino acids [26]. These components could be transformed by microbial metabolism or directly released into the fermentation system, exerting positive or negative impacts on the flavor of Huangjiu. But it is a challenge to trace the track of specific components from Tartary buckwheat to fermented Huangjiu.
2-phenoxyethanol was detected with a quite higher concentration during the Tartary buckwheat Huangjiu fermentation. Obviously, this aromatic compound possibly exerts an important effect on the flavor of Huangjiu. The presence of 2-phenoxyethanol was due to both Tartary buckwheat and microbial fermentation. Seventy-nine genera of microbes such as Corynebacterium were possible producers of 2-phenoxyethanol. To our knowledge, there is no existing report on 2-phenoxyethanol biosynthesis in microorganisms. As this compound is an important chemical material and mainly produced by chemosynthesis [41], the findings provide possible novel methods for natural 2-phenoxyethanol production by microbial fermentation.
Higher alkanol compounds are commonly rich in plants and possess important biological activities [42]. Three higher alkanol compounds were detected during the Tartary buckwheat Huangjiu fermentation. 1-hexadecanol was detected at the beginning of fermentation, and its final concentration was extremely high by the end. It is reasonable to conclude that Tartary buckwheat provided the source of 1-hexadecanol, and its generation was enhanced by subsequent microbial fermentation. The presence of high concentrations of higher alkanol compounds highlights the healthcare feature of the Tartary buckwheat Huangjiu product.
Four bacterial strains from the dominant genera and the genera correlated with 2phenylethanol and isobutanol biosynthesis during fermentation were utilized for the improved wheat Qu making and Huangjiu brewing. The validation fermentation experiments confirmed the positive correlation between specific bacterial strains and higher alcohols production. Though more validations are needed, useful cues provided by correlation analysis generate more possibilities for the development of distinguished starter cultures (Qu) to bring about an improvement in the quality of Chinese Huangjiu.