The traditionally prepared “emao” sample was collected in a sterile container from a local brewer of Rajapanichanda, Rani, Kamrup, Assam, and processed immediately for culturing fungi as described below.

The dried starter culture was crushed into powder with the help of a sterile spatula, and 10 g of it was homogenized in 90 mL of 0.1% phosphate-buffered saline (PBS) for 2–5 min from which serial dilutions (10⁻¹–10⁻⁷) were made. Three diluents, i.e., 10⁻³, 10⁻⁵, and 10⁻⁷ were used for inoculation in the Sabouraud dextrose agar (SDA) medium. In each 50 mL capacity Petri plate containing 30 mL of the agar medium, 300 μl of a diluent was spread with the help of a glass spreader. SDA medium was supplemented with tetracycline (25 μg/mL working concentration) to stop the bacterial growth. The plates were then incubated at 25 ± 1°C for 3–4 days till the appearance of the colonies. The colony-forming unit (CFU) was counted from three replica plates inoculated with a suitable dilution and the average CFU was calculated from the replicates. Distinct non-overlapping colonies were sub-cultured in fresh agar medium to get the pure cultures. The pure cultures were stored at 4°C in agar slants for routine work, and at −80°C in broth medium amended with 20% glycerol for long time preservation.

The colony and cellular morphology of all the pure culture isolates were studied to group the isolates according to their morphological similarities. Each pure culture was grown in a PDA medium incubating at 25 ± 1°C for 72 h for studying its morphological characteristics. Colony morphology was studied with the help of a stereo-zoom microscope (LM52, Lawrence and Mayo). The thallus structure was stained with lactophenol blue and observed under a trinocular compound microscope (Evos FL Cell Imaging System, Invitrogen). Both morphological and microscopic characters were compared with the characters of species described in Kurtzman and Fell (1998) for identification.

The representative fungal isolates were considered for genomic DNA isolation following the method of Cenis (1992). The nuclear DNA internal transcribed spacer (nrDNA ITS) regions were amplified in a thermal cycler (Mastercycler GSX1, Eppendorf, Germany) using the universal primer pairs: ITS1 (5′-TCCGTAGGTGAACCTGCGG-3′) and ITS4 (5′-TCCTCCGCTTATTGATATGC-3′). The polymerase chain reaction (PCR) contained 1 × PCR master-mix (Bangalore Genei), 0.4 μM each of forward and reverse primers and nuclease-free water in a 25 μl reaction volume. The PCR program was a pre-denaturation at 94°C for 3 min, 35 cycles of denaturation at 94°C for 1 min, annealing at 50°C for 30 s, extension at 72°C for 1 min, followed by a final extension at 72°C for 5 min and preservation at 4°C. The PCR products were separated in 1% agarose gel containing ethidium bromide (0.2 μg/mL) and visualized in a UV transilluminator (UVITEC, Cambridge). The amplified DNA bands were excised and extracted using QIAquick Gel Extraction Kit (QIAGEN) and were sequenced by outsourcing. The DNA sequences were deposited in the NCBI database.

The representative fungal isolates were tested for three different enzyme activities viz. amylase, cellulase, and protease in agar plates following the well-diffusion method. The amylase production and activity were tested in a starch agar medium by following the method of Stolp and Gadkeri (1981). In the case of yeasts, each pure culture isolate was inoculated at three equally distant points in an agar plate, whereas in the case of molds three separate plates were considered and inoculated with inoculum at the center of the plate to avoid the overlapping of colonies due to their vigorous growth. All the inoculated culture plates were incubated at 27 ± 1°C for 72 h and thereafter treated with Lugol's iodine solution for 15 min. Excess iodine was drained off and the culture plates were observed for the formation of a halo zone around the colonies. A colorless halo zone formation was considered amylase positive result, while blue or purple coloration was a negative result.

The cellulase activity of the isolates was tested using a CMC-Bushnell Haas Agar medium by following the method of Singh et al. (2013). The pure culture isolates were inoculated and incubated as described above for the amylase activity test. After 72 h of incubation, 0.3% of Congo red solution was poured over the culture media and kept for 20 min. Thereafter, the excess stain was drained off and the plates were washed with 1 M NaCl solution. The formation of a halo zone around the colonies indicates the cellulase positive result, while its absence indicates the negative result.

The protease enzyme activity was screened in Skim Milk Agar (SM) plates following the method of Atlas (2006). The inoculation and incubation procedure was the same as the amylase test described above. The formation of the halo zone around the colonies indicates protease positive, while the absence of the halo zone indicates the negative result.

The semi-quantitative enzymatic index was calculated for amylase, cellulase, and protease from the following equation (Coronado-Ruiz et al., 2018):

For determining the optimum growth temperature, the representative fungal isolates were inoculated in PDA plates and incubated at five different temperatures, i.e., 25, 30, 35, 40, and 45°C for 3 days. The experiments were set in triplicates for each isolate along with one negative control. The growth performance of the isolates was measured by the colony diameter size. The optimum temperature for growth was determined based on the maximum diameter of the colonies for each isolate.

To determine the optimum growth pH, each representative isolate was inoculated in three separate test tubes containing potato dextrose broth (PDB) medium adjusted to a particular pH. The pH considered were 1.5, 2, 2.5, 3, 3.5,.………..up to 10 for the experiment. A volume of 400 mL PDB was prepared and the pH was adjusted by adding either NaOH (for basic pH) or citric acid (for acidic pH). Experiments were set in 10 mL volume in triplicates for each isolate and three negative controls for each pH without adding inoculum. Actively growing pure culture isolates in PDA plates were used for inoculation in PDB tubes with the help of a sterile inoculating needle. The inoculated tubes were screw-capped and incubated at 30°C for 3 days in an orbital shaker. The tubes were centrifuged at 7,000 rpm for 15 min to collect the biomass and the supernatant was decanted off. The biomass was dried overnight in a hot air oven at 50 ± 2°C. The difference in weight of the isolates showing maximum biomass weight among the different pH was considered to have optimum growth at that pH.

The PDB medium containing different percentages of ethanol ranging from 0 to 16% was prepared for the ethanol tolerance assay of the representative isolates. The setting of experiments, inoculation, incubation, and biomass estimation of the fungal isolates was done as described above for the determination of optimum pH for growth.

The agar well diffusion method as described by Padhi et al. (2016) was followed for this experiment. To extract the crude fungal metabolites, each fungal isolate was grown in a 250 mL capacity Erlenmeyer flask containing 100 mL of sterile Potato dextrose broth (PDB) and incubated at 30°C for 7 days by shaking at 120 rpm in an incubator. The fungal culture was filtered through a Whatman filter paper 1 to remove the mycelial mass. The filtrate was taken in a separating funnel and extracted with an equal volume of ethyl acetate (99.8%) by shaking for 20 min. Two distinct phases appeared when the separating funnel was standstill for a few minutes. The lower aqueous phase containing debris was discarded and the upper phase containing ethyl acetate was collected and transferred to a heating rotary evaporator. The solution was heated at constant temperature (40°C) and pressure (239 mbar) to evaporate the solvent and the leftover dry mass was thereafter dissolved in 1 mL of 10% dimethyl sulfoxide (DMSO) and stored at 4°C for further use.

Antibacterial activity was performed against three bacterial test organisms, viz. Escherichia coli (MTCC 433), Pseudomonas aeruginosa (MTCC 2582), and Staphylococcus aureus (MTCC 9886) were procured from the Institute of Microbial Technology (IMTECH), Chandigarh, India. The test organisms were grown separately in 10 mL of nutrient broth at 35 ± 2°C for 24 h and thereafter 0.2 mL of each bacterial suspension was evenly spread over a fresh nutrient agar plate with the help of a sterile glass spreader. Three agar wells were made at equal distances in each bacterial agar plate with the help of a sterile cork-borer (8 mm in diameter). Each well was then filled with 0.1 mL of crude metabolite extracts of the fungal isolates and the plates were incubated at 35 ± 2°C for another 24 h. The antibacterial activity against the test organisms was determined based on the halo zone formations around the agar wells. DMSO (10%) was used as a negative control along with the antibacterial assay for each fungal extract.

The ethanol-producing capability of the representative fungal isolates from the glucose solution (containing 2% glucose, 1% peptone, 1% yeast extract) and starch solution (containing 2% soluble starch, 1% peptone, 1% yeast extract) was tried for all the representative isolates one by one in a liquid state fermentation. Each isolate was grown in 10 mL of sterile PDB for 12 h at 30 ± 1°C before using 100 μl of it as an inoculum for ethanol production test in 150 mL of glucose or starch solution. Each isolate was set in triplicates along with one negative control for both glucose and starch solution. The inoculated glucose and starch solutions with fungal isolates were incubated at 30 ± 1°C for 7 days; thereafter the fermented solutions were processed immediately for ethanol estimation.

The amylolytic positive fungal strains which produced ethanol from soluble starch were considered for further beer production from rice substrates. The whole experiment was performed in two steps: inoculum preparation and beer fermentation. Each representative isolate was inoculated in a 250 mL Erlenmeyer flask containing 100 mL of sterile PDB and incubated at 30 ± 1°C for 7 days to get a sufficient amount of fungal biomass for the inoculum preparation. After 1 week, the biomass was harvested by centrifugation at 7,000 rpm for 15 min in a 50 mL capacity centrifuge tube. The biomass was washed twice with sterile distilled water by centrifugation. Finally, 2 g (fresh weight) of pure culture pellet was re-suspended in 30 mL of distilled water and homogenized properly before inoculating it to the rice substrates. For beer fermentation, 180 g of dehusked glutinous rice (Oryza sativa var. glutinosa) was washed and baked in a 500 mL capacity borosilicate glass bottle (Borosil) by autoclaving at 15 lbs pressure for 15 min. The experiment was set in triplicates for each isolate. The autoclaved bottles with rice substrates were brought inside a laminar airflow hood and allowed to cool up to ~48°C. The inoculum prepared as above was inoculated to the rice substrate and mixed properly with the help of a sterile glass rod. Thereafter, the fermentation bottles were loosely capped and incubated at 30 ± 1°C for 30 days. The same procedure was followed for all the isolates, except for the isolate Mind_RF25 which took 6 months incubation period to give the beer product. The fermentation progress was monitored every two alternate days.

Rice beers produced by the amylolytic fungal isolates were recovered after 30 days of fermentation, except for Mind_RF25 which was recovered after 180 days due to its slow-growing. All the beer samples were analyzed for the following parameters.

The average CFU calculated for three agar plates inoculated with 10⁻⁵ diluents was 13.53 × 10⁷ per gram emao sample. A total of 50 fungal isolates were randomly recovered from the mixed culture plates that were sub-cultured to get the pure cultures. The pure culture isolates were segregated into 11 distinct groups based on their morphology (Table 1, Supplementary Data S1). One representative isolate from each group was considered for nrDNA ITS sequencing and DNA sequence homology search in the NCBI database. The representative isolates were identified as Candida glabrata (Cgla_RF2), Cyberlindnera fabianii (Cfab_RF37), Hyphopichia burtonii (Hbur_RF19), Mucor circinelloides (Mcir_RF48), Mucor indicus (Mind_RF25), Penicillium citrinum (Pcit_RF32), Rhodosporidiobolus ruineniae (Rrui_RF4 & Rrui_RF43), Saccharomyces cerevisiae (Scer_RF6), Saccharomycopsis fibuligera (Sfib_RF11), and Wickerhamomyces anomalus (Wano_RF3). In a relative abundance (RA) analysis of a total of 50 isolates, W. anomalus was found dominant with 24% RA followed by C. glabrata and H. burtonii with 16% RA each; whereas M. circinelloides, M. indicus, and P. citrinum were the least with only 2% RA (Figure 1).

All the yeast isolates (Cgla_RF2, Hbur_RF19, Rrui_RF4, Rrui_RF43, Scer_RF6, Sfib_RF11, and Wano_RF3), except Cfab_RF37, were showed optimum growth at 30°C, whereas all three mold isolates (Mind_RF25, Mcir_RF48, and Pcit_RF32) showed optimum growth at 35°C like Cfab_RF37 (Table 2, Supplementary Data S2). The isolates Rrui_RF4 and Rrui_RF43 did not grow beyond 30°C, whereas the rest of the yeast isolates did not grow beyond 35°C, in contrast to the molds that also grew at 45°C.

The optimum growth pH and the pH tolerance range varied among the isolates. The growth was found optimum at pH 5 ± 0.5 for Scer_RF6, Mind_RF25, and Mcir_RF48; at pH 5.5 ± 0.5 for Rrui_RF4; at pH 6 ± 0.5 for Cgla_RF2, Cfab_RF37, and Hbur_RF19; at pH 6.5 ± 0.5 for Sfib_RF11; at pH 7 ± 0.5 for Wano_RF3; and at pH7.5 for Pcit_RF32 (Table 2, Supplementary Data S3). The pH tolerance range also varied among the isolates, i.e., pH 2–7.5 in the case of Scer_RF6, pH 2–8 in the case of Rrui_RF4, and Rrui_RF43, pH 2.5–9.5, and pH 2–9.5 in the rest of the isolates. None of the isolates grew at pH 1.5 and 10.

In the ethanol tolerance assay, Scer_RF6 was found tolerant to as high as 14% (v/v) ethanol, followed by Hbur_RF19 (12%), Cgla_RF2 (11%), and Wano_RF3 (11%) in decreasing order (Table 2, Supplementary Data S4). Rrui_RF4 and Rrui_RF43 showed the least tolerance to ethanol, i.e., only up to 2% ethanol.

Of the 11 representative fungal isolates considered for amylase, protease, and cellulase activity test, five isolates (Hbur_RF19, Sfib_RF11, Mind_RF25, Mcir_RF48, and Pcit_RF32) exhibited amylase positive, only three isolates (Hbur_RF19, Sfib_RF11, and Pcit_RF32) exhibited protease positive, and only one isolate (Pcit_RF32) exhibited cellulase positive (Table 3). The amylase activity exhibited by Sfib_RF11 was the highest with an EI of 2.9 ± 0.02 among the amylase-positive isolates.

Nine of eleven isolates produced ethanol from glucose substrates with varied concentrations (Table 4). The isolates Rrui_RF4 and Rrui_RF43 did not produce ethanol from glucose substrates. The ethanol produced by Scer_RF6 was the highest (11.45 ± 0.11% v/v) followed by Hbur_RF19 (9.43 ± 007%) and Mcir_RF48 (8.05 ± 0.07%), respectively, after 1 week of liquid-state fermentation. Five of nine isolates (Hbur_RF19, Sfib_RF11, Mind_RF25, Mcir_RF48, and Pcit_RF32) produced ethanol from soluble starch in liquid state fermentation. The concentration of ethanol was observed slightly lower in the beer produced from starch than in the beer produced from glucose for all the ethanol-producing isolates from both the substrates (Table 4).

The crude metabolic extract of all the representative fungal isolates showed antibacterial activity against E. coli (MTCC 433) and S. aureus (MTCC 9886) with inhibition zones in the range of 11–32 mm and 9–35 mm, respectively, in agar well diffusion plates (Figure 2). On the other hand, only eight fungal isolates (Cgla_RF2, Hbur_RF19, Rrui_RF4, Rrui_RF43, Sfib_RF11, Mind_RF25, Mcir_RF48, and Pcit_RF32) exhibited antibacterial activity against P. aeruginosa (MTCC 2582). The antibacterial activities exhibited by Mcir_RF48, Pcit_RF32, and Hbur_RF19 to all the test organisms were noteworthy among the fungal isolates considered in this study. The three isolates (Cfab_RF37, Scer_RF6, and Wano_RF3) that showed no inhibition to P. aeruginosa (MTCC 2582) also showed weak inhibition zone to the other test organisms, E. coli (MTCC 433) and S. aureus (MTCC 9886) (Figure 2).

The pH of the rice beers was almost the same, i.e., pH 3.54 ± 0.2 in all the samples. After 30 days of rice beer fermentation, the ethanol concentration in the beer produced by Hbur_RF19 was the highest (7.3 ± 0.09% v/v) followed by Mcir_RF48, Pcit_RF32, and Sfib_RF11 (in decreasing order) (Table 4). The isolate Mind_RF25 did not produce rice beer after 30 days of the fermentation period, but after 6 months of the fermentation period, the ethanol concentration was 3.17 ± 0.06% (v/v) which was the least among the isolates. The fungal isolates produced beer containing ascorbic acids, total phenol, SADPPH, TRP, and TAC in the range of 15.26–31.1 mg/100 mL, 186.66–241.33 mg/L, 56.85–82.84%, 0.173–0.396 absorbance at A₇₀₀, and 0.174–0.364 g AAE/L, respectively (Table 5). The ascorbic acid, total phenol, and antioxidant properties of the beer produced by Hbur_RF19 were higher, except for SADPPH than the beers produced by the other four isolates. The SADPPH of the beer produced by Mcir_RF48 was the highest (82.84 ± 0.27%) followed by the beers produced by Pcit_RF32, Mind_RF25, Sfib_RF11, and Hbur_RF19 in decreasing order.

In a Pearson correlation analysis, ethanol content was found positively correlated to ascorbic acid, total phenol, total reducing power (TRP), and total antioxidant activity (TAA) of the beers (Table 6). Of the three antioxidant properties, TRP and TAA showed a positive correlation to each other, whereas SADPPH was found negatively correlated to TRP and TAA. Ascorbic acid and total phenol showed a positive correlation to each other as well as to TRP and TAA, but a negative correlation to SADPPH of the beer. The pH was found positively correlated only to SADPPH.

In a PCA biplot (Figure 3) of biochemical parameters of beer samples that produced by the fungal trains, two strains (Mcir_RF48 and Pcit_RF32) with higher SADPPH producing in rice beer appeared together in a quarter, whereas Hbur_RF19 producing higher ascorbic acid, total phenol, TRP, and TAA in beer appeared separately in another quarter of the biplot. Similarly, Sfib_RF11 and Mind_RF25 appeared separately in another quarter of the biplot. The positively correlated biochemical parameters of the beer (ascorbic acid, ethanol, total phenol, TAA, and TRP appeared together in the F1 coordinate contributing 65.83% variability, whereas the negatively correlated parameters (pH and SADPPH) appeared in the F2 coordinate contributing 20.70% variability in the biplot. In one-way ANOVA, the p-value was >0.05, thus accepting the null hypothesis that the beers produced by all the five fungal isolates had no significant difference from each other (Supplementary Data S5).

Molds, yeasts, and bacteria are the groups of microorganisms mostly reported from the traditional beer starters of the Asian region. But the microbial species reported in different beer starters are often found different from one another which might be due to differences in cultural practices and ingredients used. Besides, strain level difference within a species is a common phenomenon that encounters even in the same starter.

The fungal species C. glabrata, C. fabianii, H. burtonii, M. circinelloides, M. indicus, P. citrinum, S. cerevisiae, S. fibuligera, and W. anomalus recorded in emao were also reported earlier from other traditional beer starters viz. apop, balma, dawdim, hamei, humao, khekhrii, marcha, thempra, and thiat from India (Teramoto et al., 2002; Buragohain et al., 2013; Bharadwaj et al., 2015; Parashar et al., 2017; Sha et al., 2018; Anupama and Tamang, 2020); banh-men from Vietnam (Dung et al., 2007); daku from China (Shi et al., 2009; Zheng et al., 2012); loogpang from Thailand (Taechavasonyoo et al., 2013), and nuruk from Korea (Bal et al., 2014). However, R. ruineniae reported from emao in this study was not reported earlier. Both the strains of R. ruineniae were found non-amylolytic, non-proteolytic, non-cellulolytic, and non-ethanol producing with a low ethanol tolerance, i.e., up to 2% (v/v) only in this study (Tables 3, 4). Thus, their function in beer fermentation is unclear; rather they can be the opportunistic microbe that came by chance from the herbs used in the beer starters. It was earlier reported as an endophyte from several plant species (Santiago et al., 2017).

In ethnic beer fermentations, dozens of naturally occurring microorganisms with unknown species information generally take part in beer fermentation in contrast to the use of only the known and well-characterized microbes in beer industries. As the metabolic potentials of microbes vary greatly from species to species or strain to strain, it is necessary to characterize and understand each microbial species/strain for its industrial uses. Low-ethanol beer fortified with health-beneficial components is in demand nowadays since the beer with high ethanol content and also less or without nutritional components is injurious to health, and can cause irreparable harm to the regular consumers. The shooting taste of most of the ethnic beers is due to their moderate to high ethanol content along with high organic acids. The flavor and aroma of ethnic beer vary due to the use of different substrates, herbal ingredients, and different cultural practices.

The naturally associated microorganisms that are either maintained through the traditional practices or originated from the ingredients used might play major roles in beer fermentations as they can perform some specific activities like the production of ethanol, enzymes, organic acids, antimicrobials, vitamins, amino acids, flavor compounds, etc. However, strains vary greatly in their activity and performance as evident from several studies including the present one. For example, the amylase activity index of different fungal strains was in the range of 1.21–2.09, whereas for protease it was in the range of 1.2–1.6 among the isolates in this study (Table 3). Accordingly, the ethanol concentrations of the beer produced by the pure culture of amylolytic fungi were in the range of 3.17–7.3% (v/v) in this study. It is noteworthy to mention that the ethanol concentration of rice beer “jou” fermented with the traditional starter “emao” of the Bodo community usually contains about 12–19% (v/v) ethanol (Das et al., 2019). Earlier findings of lower ethanol production (Hashem et al., 2021) and lower ethanol tolerance (Yamaoka et al., 2014; Garcia et al., 2021) of some yeast while growing individually corroborate our present findings of the low ethanol tolerance and low ethanol production of the fungal isolates that were grown separately on rice substrates (Tables 2, 5). Besides, the presence of a good amount of ascorbic acid, total phenols, and antioxidant activities in the beer produced by the fungal strains are important from the nutritional point of view. Polyphenols contribute to antioxidant properties, health-promoting qualities, stability, and foam retention in beer (Callemien and Collin, 2009). The presence of ascorbic acid in beer increases the preservation or shelf-life of beer or wine by preventing the oxidation of beer (Jeney-Nagymate and Fodor, 2007). Therefore, immense scope lies for new product development, improvement, or value addition to the existing products through the selection and harnessing of such potential microbes having specific functions not only in beer industries but also in other food and nutraceutical industries.

S. cerevisiae is well-known for producing high ethanol concentrations and is, therefore, industrially exploited for bio-ethanol production. Wild S. cerevisiae recovered from fermented beverages can produce more than 10% (v/v) ethanol from simple carbohydrates but not directly from starch (Tesfaw et al., 2021) which is also evident from this study. But, those wild strains of S. cerevisiae can tolerate and produce higher concentrations of ethanol if they grow in association with the lactic acid bacteria (Senne de Oliveira Lino et al., 2021). Both high ethanol producing and low ethanol-producing microorganisms are reported from several ethnic beer starters. Low ethanol-producing microbes are gaining importance for low-ethanol-containing beers, in contrast to high-ethanol producing microbes that are in demand for bio-ethanol production in industries.

The ethanol produced by fungi is of economic importance to the people. In our study, we reported the ethanol-producing capacity of the amylolytic fungi and non-amylolytic fungi in the soluble starch, glucose, and rice substrate. Hbur_RF19 showed highest ethanol production (7–8% v/v) in the soluble starch (Table 4) and rice substrates (Table 5) among the amylolytic fungi followed by Mcir_RF48 (5–7% v/v), Pcit_RF32 (5–6% v/v), and Sfib_RF11 (4–5% v/v) whereas Scer_RF6 showed the highest ethanol content of 11.45% (v/v) in glucose among the non-amylolytic fungi followed by Wano_RF3 (4.16%), Cgla_RF2 (1.28%), and Cfab_RF37 (0.87%). Tsuyoshi et al. (2005) reported the presence of S. fibuligera and H. burtonii in traditional starter culture “Marcha” that has no ethanol-producing capacity. Satari et al. (2015) reported the production of ethanol (0.4 g/L) from pretreated lignocellulose by Mucor indicus. The ethanol production by Candida glabrata, Cyberlindnera fabianii, Saccharomyces cerevisiae, and Wickerhamomyces anomalus from different carbohydrate sources was reported earlier (Van Rijswijck et al., 2017; Chang et al., 2018; Atitallah et al., 2020; Vincent et al., 2021).

Antioxidants such as phenols and ascorbic acids play an important role in health-promoting factors such as scavenging free radicals, anti-inflammatory and also to the regulation of pathogens and stability of beer. In our study, the total polyphenol in the beer samples ranged from 186 to 241 mg/L, which was higher if compared to 47–60 mg/L polyphenol of Cambodian rice wine (Chay et al., 2020), 3.36–9.63 mg/L polyphenol of Brazilian beer (Moura-Nunes et al., 2016), 140–157 mg/L polyphenol of Xaj (Handique et al., 2020), and 138–148 mg/L polyphenol of Apong (Handique et al., 2020). Ambra et al. (2021) reported that the content and concentration of polyphenols depend on the use of a variety of substrates (hops or barley) for beer production. The ascorbic acid content in beer was in the range of 15.26–31.1 mg/100 mL in the present study, which is comparable to the ascorbic acid content of grape wine (15–16 mg/100 mL) and orange juice (39–41 mg/100 mL) (Pisoschi et al., 2011).

We found that all the fungal species considered in the present study were tolerant to ethanol in the range of 7–14% (v/v) ethanol, except R. ruineniae. But such ethanol tolerance did not support that all of them are good ethanol producers because the concentrations of ethanol produced by Cgla_RF2, Cfab_RF37, and Mind_RF25 were <3% (v/v). Crude metabolite extracts of low ethanol producing strains including Wano_RF3 showed lesser inhibition than the crude extracts of higher ethanol producing strains to the test organisms. In contrast, the non-ethanol producing strains Rrui_RF4 and Rrui_RF43 exhibited antibacterial activity as good as the ethanol-producing strains (Figure 2). Such fungi having antibacterial activity against pathogenic bacteria could be useful for maintaining the quality of beer starters, and thus their presence could be the quality indication in beer fermentations.

In the present study, only a subsection of yeast and mold isolates produced ethanol from rice substrates. Six yeast isolates produced ethanol from glucose, whereas only two of them produced ethanol from starch (Table 4). In contrast, all the three mold isolates (Mind_RF25, Mcir_RF48, and Pcit_RF32) produced ethanol from both glucose and starch. This is the first time to document Penicillium citrinum (Pcit_RF32) as an ethanol producing from glucose and starch substrates. Six of 11 fungal isolates were found non-amylolytic, thus are likely to depend on the amylolytic microbes when starch is used as the sole carbon source in beer fermentations. Such microbial interaction, cross-feeding, and succession are the common phenomena that occur in all-natural ecosystems including ethnic beer fermentations. Therefore, the understanding of the functional roles of each microbial species is important for process manipulation and product improvements in fermentation biology.

Certain physiological conditions of ethanol production such as Crabtree and Pasteur effects (Moulin et al., 1984; Panikov, 2011; Pfeiffer and Morley, 2014; Hagman and Piškur, 2015) are well-studied and documented in yeasts. However, such physiological study of ethanol production by filamentous fungi is limited although there are several reports of ethanol production by filamentous fungi from different substrates. For example, Mucor circinelloides is a Crabtree-positive organism and can produce ethanol in both aerobic and anaerobic conditions (Lubbehusen et al., 2004). Accordingly, M. indicus and Rhizopus oryzae can produced ethanol anaerobically from enzyme hydrolysate of pure cellulose and acid pre-treated hydrolysate of rice straw (Abedinifar et al., 2009). M. indicus (Sharifia et al., 2008), Mucor hiemalis (Esmaeili and Keikhosro, 2018), Aspergillus flavus, A. niger, A. oryzae, and A. terreus (Liaud et al., 2014) were reported to produce ethanol from glucose. Other molds such as Amylomyces rouxii, R. oryzae, R. oligosporus, R. stolonifer was reported to produce ethanol from cooked rice (Dung et al., 2006; Song et al., 2018). Fusarium oxysporum (Ruiz et al., 2007), F. verticilloides (De Almeida et al., 2013) and F. oxysporum f. sp. lycopersici (Ueng and Gong, 1982) were also reported to produce ethanol from glucose, sucrose and xylose. Therefore, detail physiological study is required to understand the mechanism of ethanol production in filamentous fungi which would help us better understand the precise mechanism of traditional beer/wine fermentations.

During beer fermentation, the growth of many pathogenic and unwanted organisms could be arrested by several factors like ethanol, pH, lack of oxygen, low nutrient availability, and antimicrobial effects (Menz et al., 2011). However, some of the undesirable and pathogenic bacteria can survive for a prolonged period at a different range of temperatures and pH (Kim et al., 2014). The presence of such unwanted microbes in beer is a concern because such microbes can spoil the beer products mostly with undesirable flavor (Vriesekoop et al., 2013).

Fungi are reported to produce various secondary metabolites or bioactive compounds which inhibit the growth of other microorganisms. In beverage fermentations, the antagonistic effect of fungi against other microbes is due to the production of antimicrobial compounds, competition for nutrients, ethanol stress, or changes in pH (Marquina et al., 2002; Gil-Rodriguez and Garcia-Gutierrez, 2020). In the antibacterial activity assay of the fungal isolates against the three bacterial test organisms (E. coli MTCC 433, P. aeruginosa MTCC 2582, and S. aureus MTCC 9886), MCir_RF48 showed the highest inhibition against the Gram-negative bacteria whereas Pcir_RF32 showed the highest inhibition against the Gram-positive bacteria in the present study. Other than molds, the yeast Hbur_RF19 showed high inhibition against E. coli and S. aureus whereas Rrui_RF43 showed high inhibition against P. aeruginosa. Fakruddin et al. (2017) reported the antibacterial activity of S. cerevisiae against E. coli (8.7 mm), S. aureus (4.9 mm), and P. aeruginosa (9.1 mm). In contrast, S. cerevisiae showed antibacterial activity against only two organisms E. coli (12 mm) and S. aureus (9 mm) in the present study. Takahashi et al. (2008) reported Penicillium citrinum from soil having no antibacterial activity against the three test organisms, but we found it showing antibacterial activity which could be due to the strain differentiation. Mohammad et al. (2020) also reported the antibacterial activity of Mucor circinelloides against E. coli (13 mm) and S. aureus (15 mm). However, reports on the antibacterial activity of yeasts are meager. The antibacterial activity of C. glabrata, C. fabianii, H. burtonii, M. indicus, M. circinelloides, S. fibuligera, and R. ruineniae against E. coli, P. aeruginosa, and S. aureus have been reported for the first time in our study. Some bacteria are also reported to grow as contaminants during fermentation which can affect the growth of beneficial yeasts and ethanol production (Thomas et al., 2001; Skinner and Leathers, 2004). Watanabe et al. (2008) reported the fermentation of ethanol with the co-culture of microbes with antimicrobial activity that inhibits the contaminants without affecting the production of ethanol. Therefore, the knowledge of the specific role of each microbial species is of paramount significance for microbial resource management and utilization in fermentation biology.