In this study, high-throughput sequencing and headspace solid-phase microextraction (HS-SPME) coupled with gas chromatography-mass spectrometry (GC-MS) were applied to investigate the composition of acetic acid-related fungi in fermented grains for nongxiangxing baijiu in northern China and their distribution characteristics in the brewing environment during fermentation (from Days 7 to 67) in different seasons (spring and summer). The results indicated that the content of acetic acid in fermented grains increased with fermentation time and was significantly higher in summer than in spring (P < 0.05). Correlation analysis revealed that 122 genera of fungi were significantly correlated with acetic acid synthesis (P < 0.05). Among them, Thermomyces, Issatchenkia, Zygosaccharomyces, Kazachstania and Rhizopus were the dominant genera. Thermomyces and Zygosaccharomyces were positively correlated with acetic acid (P < 0.05), while Issatchenkia and Kazachstania showed significant negative correlations with acetic acid (P < 0.05). In addition, complex correlations existed between the acetic acid-related fungi and the dominant fungi. Source tracking results indicated that the fungi in the fermented grains originated from the ground (20.3%), tools (20.3%), pit mud (18.0%), Daqu^L (16.2%), raw materials (16.2%) and Daqu^F (9.1%). Moreover, six dominant genera of fungi significantly correlated with acetic acid were found in samples from both the ground and tools. In the fermented grains, Kazachstania, Thermomyces and Issatchenkia originated from pit mud, Issatchenkia from Daqu^L, Zygosaccharomyces from raw materials, and Rhizomucor and Rhizopus from Daqu^F. This study provides a theoretical basis for establishing strategies to regulate the content of acetic acid in fermented grains for Nongxiangxing Baijiu during the fermentation process.

In this study, we investigated the bacterial community composition and structure during the fermentation process of traditional light-flavor baijiu using PacBio SMRT sequencing technology, and further explored the mechanism of bacterial community self-assembly during the fermentation process using a framework to quantitatively infer community assembly mechanisms by phylogenetic bin-based null model analysis (iCAMP). The results showed that the bacterial diversity decreased and then increased during the fermentation process. Lactobacillus, Acetobacter, Pediococcus and Weissella were identified as the dominant genera, and L. pontis, L. paralimentarius, L. brevis and L. acetotolerans as the dominant species. According to the differences among bacterial communities at different fermentation times, the whole fermentation process could be divided into three stages: Ⅰ (0 days), Ⅱ (3–10 days) and Ⅲ (12–28 days). Acetobacter malorum, L. paralimentarius and L. pontis were the differential species in stages Ⅰ, Ⅱ and Ⅲ, respectively, and there were a significant negative correlation between the differential species in different stages and a significant positive correlation between the differential species in the same fermentation stage. In addition, turnover was the major process responsible for bacterial community changes, and drift during stoichiometry was the most important process driving bacterial community self-assembly. Meanwhile, starch content, temperature and moisture content may be the main factors influencing bacterial community self-assembly during fermentation. The above results provide insight into the bacterial species composition, community succession patterns and self-assembly mechanisms during the fermentation of traditional light-flavor baijiu, which can provide a theoretical basis for regulating the microbial communities in the baijiu brewing process and further optimizing the role of multiple microorganisms in the mixed fermentation process.