Effects of Five Different Lactic Acid Bacteria on Bioactive Components and Volatile Compounds of Oat

In this research, oats were fermented with Lactobacillus plantarum, Lactobacillus acidophilus, Lactobacillus casei, Lactobacillus bulgaricus and Streptococcus thermophilus for 48 h at 37 °C. The purpose of this work was to compare the growth capacities of the five lactic acid bacteria (LAB) in the oat matrix and the effects of fermentation on the contents of the bioactive components of oat, such as β-glucan, polyphenols, flavonoids and volatile compounds at different time (0, 4, 8, 12, 24, 36 and 48 h). After 48 h of fermentation, the number of living L. acidophilus in oat reached 7.05 × 109 cfu/mL, much higher than that of other strains. S. thermophilus retained the greatest β-glucan content, and L. casei had increased total polyphenol and total flavonoid contents. The proportion of free and bound polyphenols and flavonoids in all samples was changed by microbial action, indicating that forms of polyphenols and flavonoids can be transformed during the fermentation process, and the changes varied with different strains. The samples with L. plantarum, L. acidophilus, and L. casei fermentation contained more alcohols, whereas those with S. thermophilus and L. bulgaricus fermentation had more aldehydes, which revealed that the composition of volatile components was related to strains. The results indicate that oat substrate is a good medium for LAB growth. This study provides a reference for the use of different strains to achieve different fermentation purposes and a theoretical basis for the further processing of oat and fermented oat beverages.


Introduction
Oats are grown all over the world, rank seventh in global production behind corn, wheat, rice, barley, sorghum, and millet [1], and are rich in carbohydrates, balanced protein, essential fatty acids, vitamins, and other nutrients [2]. In addition, oats contain many bioactive substances, such as β-glucan, polyphenols, and flavonoids. Among them, βglucan is a dietary fiber with hypoglycemic and lipid-lowering effects, which can reduce the risk of obesity, diabetes, and cardiovascular diseases and is of great benefit to human health [3,4]. New research shows that polyphenols can improve intestinal health and plasma inflammation and participate in cell signal transduction pathways owing to their anti-inflammatory, antithrombotic, and antioxidant activities [5]. The oat-based food industry has been promoted by the U.S. Food and Drug Administration's recommendation to eat more oats because of their high nutritional benefits and the growing emphasis on healthy food [6]. Thus, exploring more oat processing technologies, enriching oat products, and improving economic benefits are the directions of future development.
Ways to improve the nutritional value of grains include cooking, grinding, and fermentation [7]. Bioactive substances in the grain bran tend to bind to the complex structure of the cell wall, which resists traditional crushing processes; however, bioprocessing techniques can effectively solve this problem [8]. Fermentation may be most economical and the simplest way to increase the nutritional value and functional quality of oats [9]. Lactic acid 2.5. Determination of β-Glucan Content β-Glucan content was evaluated by a β-glucan assay kit (mixed linkage) (Megazyme, Wicklow, Ireland). A volume of 3 mL of sample was added to 9 mL of 95% ethanol and stood for 5 min. After the supernatant was discarded (1000× g, 10 min; Sigma, Lower Saxony, Germany), 10 mL of 50% ethanol was added, dispersed again and the precipitate reserved (1000× g, 10 min). Volumes of 0.2 mL of 50% ethanol and 4 mL of sodium phosphate (20 mM, pH 6.5) were added to the precipitate followed by heating in a boiling water bath for 3 min. The tube at was incubated at 50 • C for 5 min and 0.2 mL of lichenase solution (50 U/mL) was added. The reaction was carried out for 1 h with vigorous stirring (3-4 times). After that, 5 mL sodium acetate buffer (200 mM, pH 4.0) was added and the tube allowed to equilibrate to room temperature. The supernatant after centrifugation (0.1 mL; 1000× g, 10 min) was hydrolyzed with 0.1 mL β-glucosidase solution, and the obtained solution was used for color reaction.

Extraction of Phenolic Compounds
The extraction of phenolic compounds was performed according to the method described by Zhang et al. [17] with some modifications. O-Lp, O-La, O-Lc, O-Lb, and O-St samples (4 mL) with different fermentation time were collected, mixed with 20 mL of 80% ethanol, and placed in an ultrasonic cleaner (KQ-50E, Kun Shan Ultrasonic Instruments Co., Ltd., Kun Shan, China) at 25 • C for 20 min. The obtained liquid was centrifuged at 4000 rpm for 10 min at 4 • C and the above steps repeated three times. The supernatant was collected and concentrated with a vacuum rotary evaporator (RV 10 digital V, IKA, Baden-Wuerttemberg, Germany) at 40 • C. The liquid, containing free phenolics, was kept at a constant volume of 10 mL with methanol and stored under dark condition. The remaining precipitate was added to hexane to remove lipids. Then, the precipitate was hydrolyzed by addition of 20 mL of 4 M NaOH, shaken for 1 h, and adjusted to pH 2.0-3.0 with 6 M HCl. The mixture was extracted with 20 mL of ethyl acetate, followed by ultrasonication for 20 min and centrifugation at 4000 rpm for 10 min. The operation was repeated three times to collect the supernatant, which was vacuum-evaporated at 40 • C. The liquid, containing the bound phenolics, was kept at a constant volume of 10 mL with methanol and stored under dark conditions.

Determination of Phenolic Content
Phenolic content was determined using an adapted and validated method [18] with slight modifications. Briefly, 0.25 mL of extract was mixed with 1 mL of distilled water and 0.25 mL of Folin-Ciocalteu's phenol reagent was added for reaction for 6 min. Then, 2.5 mL of 7% Na 2 CO 3 and methanol were added to a total volume of 10 mL. The obtained liquid was incubated for 90 min in darkness at room temperature. Subsequently, absorbance was measured on a spectrophotometer (UV-2100, UNICO, Shanghai, China) at 760 nm. The phenolic compounds were quantified using a gallic acid standard calibration curve. The results were expressed in milligram of gallic acid equivalent (GAE) per 1 L of sample.

Determination of Flavonoid Content
Flavonoid content was measured with reference to the method proposed by Kim et al. [19]. A 1 mL of sample was diluted to 5 mL with 70% ethanol and mixed with 0.3 mL of 5% NaNO 2 for 5 min. Afterward, 0.3 mL of 10% AlCl 3 ·6H 2 O was added to react for 6 min. Finally, 2 mL of 1 M NaOH and 2.4 mL of 70% ethanol were added, and the obtained liquid was incubated for 15 min. The absorbance was read at 420 nm. A rutin standard calibration curve was used for the quantification of flavonoids, and the results were expressed as milligram of rutin equivalent (RE) per 1 L of sample.

Determination of Volatile Components
Samples were incubated at 50 • C for 10 min in a headspace flask (20 mL), extracted continuously with fiber (DVB/CAR/PDMS, 50/30 µm) at 50 • C for 30 min, and desorbed at 250 • C for 5 min.
Two-dimensional gas chromatography-time-of-flight mass spectrometry analysis was conducted using a Pegasus GC-HRT+ 4D high-performance mass spectrometer (LECO Corp., San Jose, CA, USA). Separation was carried out on a MAT-WAX column (30 m × 0.25 mm × 0.25 µm film thickness; Restek, Bellefonte, PA, USA) with a helium (purity > 99.999%) carrier gas, maintaining a constant flow rate of 1 mL/min. The splitless mode was operated. The inlet temperature was set at 250 • C. The oven temperature was initially set at 40 • C for 3 min, then programmed to increase to 230 • C at a rate of 10 • C/min, and held at 230 • C for 6 min. The transmission line temperature and ion source temperature were set to 250 • C. Mass spectra were measured over a range of 33-400 m/z utilizing an electron energy of 70 eV.

Statistical Analysis
All experiments were conducted in triplicate unless specified. Diagrams were drawn using Origin 2022b (Origin Statistical Software, Northampton, MA, USA). ANOVA with Duncan's test (p < 0.05) was used to analyse the differences between samples in SPSS 24.0 statistics software (SPSS Inc., Chicago, IL, USA).

LAB Growth Curve
The purpose of this work was to monitor and compare the cell viabilities of L. plantarum, L. acidophilus, L. casei, S. thermophilus, and L. bulgaricus inoculated in oats. As shown in Figure 1, the five strains showed remarkable differences in growth activity during the first 12 h of the fermentation. The differences were caused by the strains' utilization of carbon and nitrogen sources and adaptability to substrates [20]. L. acidophilus and L. plantarum had the fastest growth rates. The maximum concentrations of all strains except L. bulgaricus were determined after inoculation for 24 h. At this time, L. acidophilus entered the stable stage, and the number of viable bacteria reached 1.36 × 10 10 colony forming units (cfu)/mL. The results showed that L. acidophilus and L. plantarum had stronger growth abilities in oat substrate than the other strains. The viable counts of the five strains in the lag phase were all higher than the recommended minimum of 10 6 cfu/mL for probiotic products [21], indicating that all strains could use the nutrients in oats, and the oat substrate was a suitable fermentation substrate. Similar results were shown by Rathore et al. [22]. Who found that L. plantarum and L. acidophilus grew well in barley and malt substrates. In fact, L. plantarum, L. acidophilus, and L. casei are considered common strains for cereal fermentation [12]. LAB are heterotrophic organisms that lack some biosynthetic processes and have complex nutritional needs [23]. Different LAB can metabolize different carbon sources, and each microorganism shows a specific preference for one or more sugars [24]. This is also why L. bulgaricus and S. thermophilus have poorer growth abilities in oat compared with other strains, and prefer to use lactose.

Effect of Fermentation on pH
The ability of LAB to produce acid during fermentation is related to whether the growth of miscellaneous bacteria can be inhibited during product storage, and also affects the sensory characteristics of the final product. The results revealed that the five strains could reduce the pH of the substrate to varying degrees during fermentation (Table 1) and finally reach a pH of 3.15-4.25. After 8 h of fermentation, the pH values of L. plantarum and L. acidophilus decreased rapidly by 27.39% and 27.67%, respectively, which were distinctly different from those of other strains (p < 0.05). The maximum decrease in the pH value of the two strains occurred at the same time with the index growth period of the strain. By contrast, S. thermophilus has the weakest acid-producing capacity, reaching a pH of 4.25 after 48 h of fermentation, which may be attributed to L. acidophilus preferring lactose to glucose as its main energy source [25]. LAB can produce lactic acid through the carbohydrate metabolism pathway. The production of various organic acids may be the main reason for the decrease in pH [26]. The above results are supported by the study of Mirmohammadi et al. [27]. They concluded that the pH values of different substrates decrease rapidly within 12 h after fermentation by LAB. The strains used in the production of fermented cereal beverages and fermentation time affect the pH of the product.

Effect of Fermentation on pH
The ability of LAB to produce acid during fermentation is related to whether the growth of miscellaneous bacteria can be inhibited during product storage, and also affects the sensory characteristics of the final product. The results revealed that the five strains could reduce the pH of the substrate to varying degrees during fermentation (Table 1) and finally reach a pH of 3.15-4.25. After 8 h of fermentation, the pH values of L. plantarum and L. acidophilus decreased rapidly by 27.39% and 27.67%, respectively, which were distinctly different from those of other strains (p < 0.05). The maximum decrease in the pH value of the two strains occurred at the same time with the index growth period of the strain. By contrast, S. thermophilus has the weakest acid-producing capacity, reaching a pH of 4.25 after 48 h of fermentation, which may be attributed to L. acidophilus preferring lactose to glucose as its main energy source [25]. LAB can produce lactic acid through the carbohydrate metabolism pathway. The production of various organic acids may be the main reason for the decrease in pH [26]. The above results are supported by the study of Mirmohammadi et al. [27]. They concluded that the pH values of different substrates decrease rapidly within 12 h after fermentation by LAB. The strains used in the production of fermented cereal beverages and fermentation time affect the pH of the product.

Effect of Fermentation on β-Glucan Content
Our work aimed to study the effect of LAB on β-glucan content during oat fermentation. The β-glucan contents of the five strains shown in Figure 2 increased at the initial fermentation stage, which may be related to the fact that insoluble β-glucan was degraded to soluble β-glucanase by LAB [28], and then LAB consumed a large amount of carbohydrates for proliferation. β-Glucan can also provide growth substrates (prebiotics) for some LAB. Therefore, the contents of all samples were significantly decreased at 8-12 h of fermentation (p < 0.05), but the decreases were different. The sample fermented with S. thermophilus for 24 h decreased the β-glucan content by 5.09% compared with the unfermented sample; thus, O-St had the most β-glucan content. Evidence shows that S. thermophilus TKM3 KKP2030p does not grow well in oat-banana matrix and does not utilize β-glucan [29]. This finding was supported by the results of the viable count experiment and β-glucan content in the present study. Studies showed that the change in β-glucan content during fermentation is related to the strain type. Interestingly, Sims et al. [30] concluded that β-glucan oligosaccharide supported L. rhamnosus growth, but B. lactis and L. acidophilus did not grow on this substrate. Therefore, the different utilization of β-glucan by different strains in the oat matrix may be the reason for the inconsistent decrease in β-glucan content. These data provide guidance for the development of fermented oat beverages. Strains that do not ferment β-glucan can be selected to maximize the potential of probiotics.

Effect of Fermentation on β-Glucan Content
Our work aimed to study the effect of LAB on β-glucan content during oat fermentation. The β-glucan contents of the five strains shown in Figure 2 increased at the initial fermentation stage, which may be related to the fact that insoluble β-glucan was degraded to soluble β-glucanase by LAB [28], and then LAB consumed a large amount of carbohydrates for proliferation. β-Glucan can also provide growth substrates (prebiotics) for some LAB. Therefore, the contents of all samples were significantly decreased at 8-12 h of fermentation (p < 0.05), but the decreases were different. The sample fermented with S. thermophilus for 24 h decreased the β-glucan content by 5.09% compared with the unfermented sample; thus, O-St had the most β-glucan content. Evidence shows that S. thermophilus TKM3 KKP2030p does not grow well in oat-banana matrix and does not utilize β-glucan [29]. This finding was supported by the results of the viable count experiment and β-glucan content in the present study. Studies showed that the change in β-glucan content during fermentation is related to the strain type. Interestingly, Sims et al. [30] concluded that β-glucan oligosaccharide supported L. rhamnosus growth, but B. lactis and L. acidophilus did not grow on this substrate. Therefore, the different utilization of β-glucan by different strains in the oat matrix may be the reason for the inconsistent decrease in β-glucan content. These data provide guidance for the development of fermented oat beverages. Strains that do not ferment β-glucan can be selected to maximize the potential of probiotics.

Effect of Fermentation on Phenolic Content
In fermented products, the glycoside form of polyphenols can be transformed into the aglycone form by microorganisms to improve their bioavailability in the intestine and perform their beneficial functions better. Changes in total phenolic content (TPC), bound phenolic content (BPC), and free phenolic content (FPC) during oat fermentation by LAB are depicted in Figure 3a-c, respectively. The results indicate that a strain has strict specificity in phenolic acid metabolism/degradation/hydrolysis as reported in the literature [31]. For example, as shown in the figure, the TPCs of some strains were distinctly different at the given fermentation time (p < 0.05). Compared with the unfermented samples, the TPCs of all fermented samples showed a decreasing trend at the later stage of fermentation, particularly after 12 h, except for O-Lc. However, after 48 h of fermentation, only

Effect of Fermentation on Phenolic Content
In fermented products, the glycoside form of polyphenols can be transformed into the aglycone form by microorganisms to improve their bioavailability in the intestine and perform their beneficial functions better. Changes in total phenolic content (TPC), bound phenolic content (BPC), and free phenolic content (FPC) during oat fermentation by LAB are depicted in Figure 3a-c, respectively. The results indicate that a strain has strict specificity in phenolic acid metabolism/degradation/hydrolysis as reported in the literature [31]. For example, as shown in the figure, the TPCs of some strains were distinctly different at the given fermentation time (p < 0.05). Compared with the unfermented samples, the TPCs of all fermented samples showed a decreasing trend at the later stage of fermentation, particularly after 12 h, except for O-Lc. However, after 48 h of fermentation, only the TPC of O-Lc increased by 3.14%. Similar results were found by Li et al. [32], who fermented jujube juice with L. plantarum and L. casei to increase TPC. Other results showed that fermentation could also reduce phenolic content. Moreover, extractability was reduced by the selfpolymerization of phenolic compounds and/or the interaction with other macromolecules (such as amino acids and starch) [33]. In addition, compounds are also be transformed and degraded into other healthy monomers [8].  Figure 3b,c show that the FPCs of the five strains were higher than that of the control sample within 24 h of fermentation, whereas the BPCs had the opposite trend. Additionally, an increase in the FPC of each sample was often accompanied by a decrease in BPC. This phenomenon can be explained by the transformation of BPC into FPC due to the metabolic activities of microorganisms. LAB grow rapidly in the early stage of fermentation and can use sugar and protein to partially release BPC. In addition, esterase, decarboxylase, and β-glucanase are produced in the proliferation process. Among them, β-glucanase hydrolyzes the β-glycosidic bond of conjugated phenolic compounds [34], resulting in the release of conjugated phenolic compounds and an increase in FPC [35]. Ferulic acid esterase also releases BPC from the grain cellulose matrix into a free form. At different fermentation times, the FPC of O-Lc was always higher than those of other samples (p < 0.05), especially at 12 h, when it could reach 73.70 mg GAE/L, which was 1.32 times higher than that of the control samples. The BPC in O-Lc was remarkably lower than those of O-St, O-Lc, and O-Lb. In addition to glucanase activity, the ability of bacterial strains to degrade phenol esters or tannins, different induced phenol decarboxylase activities, and different substrate acidity may also contribute to this result [36]. Overall, fermentation can change the phenolic contents in the samples. However, whether the compositions of free and bound monomeric phenolic compounds change needs further verification.

Effect of Fermentation on Flavonoid Content
Flavonoids are important natural organic compounds that exist widely in nature. Most flavonoids have strong biological activities, which has aroused research interest. Flavonoids occupy an increasingly important position in daily diet and disease treatment owing to their extensive pharmacological effects and low toxicity. The content change of flavonoids during oat fermentation is shown in Figure 4. The content of total flavonoids in oats fermented by different strains showed a downward trend within 48 h as shown in Figure 4a. This result may be related to the large reduction of conjugated flavonoids. According to previous reports, flavonoids can be transformed into free forms in the fermentation process of soybean and tea, but a decrease in the total amount may not indicate a decrease in biological activity. Bound and free flavonoids decreased during the fermentation of corn and other grains [37]. Researchers attributed this to the metabolism of these compounds during microbial fermentation, such as the degradation/polymerization of flavonoids into dihydroxy flavonoids analogues and the activity changes of enzymes, including glycosidase, glycosyltransferase, tannase, esterase, and hydrolase [34]. In addition, some reactions may induce flavonoids to transform into other metabolites through flavonoid methylation, glycosylation, flavonoid deglycosylation, and flavonoid-sulfuric acid conjugation. Figure 4b,c shows that after 4 h of fermentation, the free flavonoids in the samples of the other four bacteria except O-Lp were increased, and the bound flavonoids were decreased. O-Lp had the opposite trends initially but gradually had the same trends in the later fermentation time. This outcome could be related to the depolymerization of bound flavonoids and the formation of soluble free flavonoids. Interestingly, we found that at each fermentation time point, the free flavonoid content of O-Lc was distinctly higher than in those of other strains (p < 0.05), reaching 202.60 mg RE/L at 48 h, an increase of 20.47% compared with the unfermented sample. Flavonoid glycosides may be consumed during fermentation and released in the form of aglycones. LAB can convert flavonoid glycosides in Cudrania tricuspidate leaves into flavonols, quercetin, and kaempferol [38].
Current studies on the biotransformation of flavonoids in food fermentation processes are few. Xu et al. [39] fermented milk containing Scutellaria baicalensis with Lactobacillus brevis and found that baicalin and wogonoside were converted into their aglycone forms, baicalein and wogonin, respectively, which have higher biological activities. The biotransformation mechanism of flavonoids in LAB-fermented food remains to be further studied.

Effect of Fermentation on Volatile Components
The volatile components of oat fermented by different LAB strains for 48 h are shown in Table 2. The volatile components included 10 alcohols, 10 aldehydes, 7 acids, 15 ketones, 7 esters, 11 furan derivatives, 8 hydrocarbons, and 1 terpene. Oat is prone to oxidative rancidity and deterioration during processing, storage, and circulation, which is related to the high fat content of oat, especially the high percentage of unsaturated fatty acids and the large amount of lipase with high activity in the endosperm. The oxidative cleavage of oleic or linoleic acid during the contact between unsaturated fatty acids and lipase in oat crushing or milling produces hexanal and nonanal. Nonanal production may be caused by the loss of a hydrogen from the 10th carbon of the oleic acid chain, followed by the absorption of a hydrogen peroxyfree group (OOH) and subsequent fracture. The cleavage of the 13-hydroperoxide after the oxygenation of linoleic acid chain may lead to the formation of hexaldehyde. The fatty oxygenase-specific oxidation of 9-hydroperoxides can form 2-pentylfurans, which have greeny, beany, and buttery aromas, whereas 1-octen-3-ol may arise from the 10-hydroperoxide of linoleic acid. This finding was confirmed by the fact that 1-octene-3-ol, hexanal, and nonanal had high concentrations in the unfermented samples as presented in Table 2, revealing that partial oxidation occurred before fermentation. At the same time, we observed that 2-pentylfuran, as one of the important volatile compounds, was detected in each fermented sample, and was significantly higher than that in the fermented sample (p < 0.05). This indicates that oxidation also occurred during fermentation.
No pentanal, hexanal and heptanal were detected in O-Lc and O-La. This result was very similar to the conclusion of Lee et al. [40], who fermented oats with L. paracasei and found that hexanal content decreased considerably within the first 2 h and was completely undetectable at 24 h. This outcome was due to the fact that microbial action can convert aldehydes into alcohols and acids. In our experiment we observed the fact that 1-pentanol, 1-hexanol, 1-heptanol, acetic acid and hexanoic acid increased correspondingly. Overall, after LAB fermentation, the contents of aldehydes in the sample decreased, and the contents of alcohols, acids, and ketones increased. However, there were distinct differences in flavor components among fermented samples with different strains. As can be seen from Table 2 Fermented oats contain volatile components, such as 1-hexanol, hexanal, nonanal, acetic acid and 2-pentylfuran. These components are also considered the key flavor compounds in Lactobacillus-fermented foods and influence the organoleptic properties of fermented products.

Conclusions
In summary, oat substrate is a suitable fermentation substrate. L. plantarum, L. casei, L. acidophilus, L. bulgaricus, and S. thermophilus could grow well in oat substrate, and the number of viable bacteria could reach 10 6 cfu/mL even at the late fermentation stage. L. acidophilus showed the strongest growth ability in oat substrate, and the number of live bacteria was the highest. Soaking oats for 1 h prior to fermentation made them more conducive to subsequent cleaning and absorption of water without significantly affecting the bioactive components. L. plantarum and L. acidophilus had the strongest acidproducing capacity during the fermentation process, and S. thermophilus retained the most β-glucan content after 48 h of fermentation. Moreover, the contents of total polyphenols and total flavonoids in oats varied with different strains, among which O-Lc was the highest. Aldehydes were predominant in O-Lb and O-St, but alcohols were predominant in O-Lp, O-Lc, and O-La. In conclusion, different strains used in oat fermentation have different effects. Therefore, suitable LAB can be selected in future research to improve beneficial ingredients through microbial-mediated biotransformation, providing guidance for the research and development of cereal fermentation products and other options for vegetarians and lactose-intolerant people.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.