Fermentation agent for preparing special flavor and storage resistant milk bean curd and application thereof

Abstract

The application discloses a kind of special flavor and storage-resistant milk bean curd fermentation inoculant and application thereof, belong to the field of microbial technology, provide a kind of special flavor and storage-resistant milk bean curd fermentation inoculant, including pentose lactobacillus plantarum ( Lactiplantibacillus pentosus ) YN-04 and lactobacillus plantarum ( Lactiplantibacillus plantarum ) YN-05, on March 13, 2025, preserved in China typical culture preservation center, and the preservation address is China. Wuhan. Wuhan University, and the preservation number is CCTCC NO: M2025469 and CCTCC NO: M2025470 respectively. The fermentation inoculant of the application can synthesize enzymes promoting special flavor and beneficial storage, the enzyme types synthesized by the two bacteria are complementary, and have excellent fermentation characteristics.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of microbial technology, and particularly relates to a fermentation inoculant for preparing special-flavor and storage-resistant milk bean curd and application thereof. BACKGROUND

[0002] Probiotics play a crucial role in maintaining the health of the body. After being ingested into the host, probiotics can improve the structure of intestinal flora, promote the proliferation of beneficial bacteria in the intestine, inhibit the growth of harmful bacteria, and enhance the specific or non-specific immunity of the body, thereby helping to resist various diseases. In recent years, probiotics have shown broad application prospects in food processing, pharmaceutical and healthcare, and livestock breeding. Probiotic preparations have wide application value in agriculture, food and medical fields. Traditional bacterial agents generally have low survival rate (survival rate < 50% at room temperature) and single function.

[0003] Pentosus lactis is a lactic acid bacterium with special metabolic characteristics. It can utilize a variety of carbon sources for fermentation and produce organic acids, antibacterial substances and other beneficial metabolites. Studies have shown that P. lactis not only has inhibitory effect on intestinal harmful bacteria, but also can promote the absorption of nutrients, thus having important application value in functional food and feed additives. Lactiplantibacillus pentosus Lactobacillus plantarum is one of the typical representatives of the genus Lactobacillus and widely exists in fermented foods in nature. This bacterium has strong tolerance and can survive in the gastrointestinal environment and play a probiotic role. In addition, L. plantarum can secrete antioxidant substances and degrade anti-nutritional factors, thereby improving the health level of the host. Lactiplantibacillus Lactiplantibacillus plantarum P. lactis and L. plantarum, as resident beneficial microorganisms in the gastrointestinal tract, play a key role in stabilizing and maintaining the intestinal microecological environment. A large number of clinical studies, animal experiments and in vitro cell experiments have shown that these two bacteria not only can produce lactic acid, acetic acid and other organic acids to lower the pH value in the intestine, effectively inhibit the growth of harmful bacteria such as Escherichia coli and Salmonella, and create a suitable living environment for beneficial bacteria; but also can stimulate the intestinal immune system, enhance the intestinal mucosal barrier function, promote the repair and renewal of intestinal epithelial cells, and thus maintain the normal physiological metabolism and digestive absorption function of the intestine. In addition, they also have positive performance in improving lactose intolerance symptoms and assisting in regulating blood lipid and blood glucose levels.

[0004] P. lactis and L. plantarum, as resident beneficial microorganisms in the gastrointestinal tract, play a key role in stabilizing and maintaining the intestinal microecological environment. A large number of clinical studies, animal experiments and in vitro cell experiments have shown that these two bacteria not only can produce lactic acid, acetic acid and other organic acids to lower the pH value in the intestine, effectively inhibit the growth of harmful bacteria such as Escherichia coli and Salmonella, and create a suitable living environment for beneficial bacteria; but also can stimulate the intestinal immune system, enhance the intestinal mucosal barrier function, promote the repair and renewal of intestinal epithelial cells, and thus maintain the normal physiological metabolism and digestive absorption function of the intestine. In addition, they also have positive performance in improving lactose intolerance symptoms and assisting in regulating blood lipid and blood glucose levels.

[0005] Despite significant progress in probiotic research, including a relatively in-depth understanding of the basic characteristics and some functions of *Lactobacillus pentosus* and *Lactobacillus plantarum*, many gaps remain in research on specific strains. While *Lactobacillus pentosus* and *Lactobacillus plantarum* are recognized as safe microorganisms and have been used in fermented foods, their synergistic effects and mechanisms have not yet been studied. Summary of the Invention

[0006] To address the aforementioned technical problems, this invention proposes a fermentation agent for preparing milk tofu with a special flavor and good storage properties, and its application. The fermentation agent includes Lactobacillus pentosus YN-04 and Lactobacillus plantarum YN-05, which can synthesize enzymes that promote special flavor and are beneficial for storage. The types of enzymes that the two can synthesize are complementary, resulting in excellent fermentation characteristics and demonstrating potential application value in the fermentation industry.

[0007] To achieve the above objectives, the present invention provides a fermentation agent for preparing milk tofu with a special flavor and good storage resistance, wherein the fermentation agent includes *Lactobacillus pentosaceus* (…). Lactiplantibacillus pentosus YN-04 and Lactobacillus plantarum ( Lactiplantibacillus plantarum YN-05;

[0008] The Lactobacillus pentosaccharide YN-04 was deposited at the China Center for Type Culture Collection (CCTCC) on March 13, 2025, at Wuhan University, Wuhan, China, with accession number CCTCC NO: M2025469.

[0009] The *Lactobacillus plantarum* YN-05 was deposited on March 13, 2025, at the China Center for Type Culture Collection (CCTCC), Wuhan University, Wuhan, China, with accession number CCTCC NO: M2025470.

[0010] Preferably, the mass ratio of *Lactobacillus pentosus* YN-04 to *Lactobacillus plantarum* YN-05 in the fermentation agent is 1:1; and the effective viable count of *Lactobacillus pentosus* YN-04 in the fermentation agent is ≥10. 9 CFU / mL, the effective viable count of *Lactobacillus plantarum* YN-05 in the fermentation agent is ≥10. 8 CFU / mL.

[0011] The present invention also provides the application of the fermentation agent in the preparation of fermented products with special flavor.

[0012] Preferably, the fermented product is milk tofu.

[0013] The present invention also provides the application of the fermentation agent in the preparation of storable fermented products.

[0014] Preferably, the fermented product is milk tofu.

[0015] The application also provides a method for preparing milk bean curd by using the fermentation agent, comprising the following steps: sterilizing raw cow milk, inoculating the fermentation agent into the sterilized raw cow milk, fermenting until the pH is 4.6, heating and draining whey, kneading, cutting, and compression molding to obtain the milk bean curd.

[0016] Preferably, the sterilization of the raw cow milk is performed by placing the raw cow milk in a water bath at 80 DEG C for 15 min, followed by cooling.

[0017] Preferably, the inoculation amount of the fermentation agent is 3% of the volume fraction of the sterilized raw cow milk; and the fermentation temperature is 25 DEG C.

[0018] Preferably, the heating and draining of whey is performed by water bath at 65 DEG C.

[0019] Compared with the prior art, the application has the following advantages and technical effects:

[0020] The application provides a fermentation agent for preparing special-flavor and storage-resistant milk bean curd, comprising Pediococcus pentosaceus YN-04 and Pediococcus plantarum YN-05. The Pediococcus pentosaceus YN-04 and the Pediococcus plantarum YN-05 have good gastrointestinal fluid tolerance, bile salt tolerance, and excellent fermentation characteristics. The Pediococcus pentosaceus YN-04 and the Pediococcus plantarum YN-05 have strong protein decomposition and amino acid synthesis capacity, can significantly form proline, creatine, L-threonine, glycine, pyruvic acid, L-valine, L-isoleucine, succinic acid and other amino acids after preparing the milk bean curd, and have significant bacteriostatic effect. The two have complementary relationship in the types of synthesis enzymes. The Pediococcus plantarum YN-05 does not have acetate kinase, phosphoacetyltransferase and ethanol dehydrogenase, but has rich amino peptidase and amino acid synthesis enzyme, and can generate various amino acids. The Pediococcus pentosaceus YN-04 has the related enzymes. The Pediococcus plantarum YN-05 contains L-lactic acid dehydrogenase and malate / lactic acid dehydrogenase. The complementary relationship of the types of synthesis enzymes of the two can improve the special flavor of the prepared milk bean curd and facilitate the storage thereof. BRIEF DESCRIPTION OF DRAWINGS

[0021] In order to more clearly illustrate the technical solutions in the embodiments of the application or the prior art, the following will briefly introduce the drawings needed in the embodiments. Obviously, the drawings in the following description only some embodiments of the application, and for those skilled in the art, other drawings can also be obtained without creative labor on the basis of these drawings.

[0022] Figure 1 The phylogenetic tree of Pediococcus pentosaceus YN-04 is shown in the figure;

[0023] Figure 2 Cholesterol standard curve;

[0024] Figure 3 Figure 2 is a base composition distribution graph and a length distribution graph of second-generation sequencing reads, wherein A is a base composition distribution graph of second-generation sequencing reads of Lactiplantibacillus plantarum YN-05, B is a base composition distribution graph of second-generation sequencing reads of Lactiplantibacillus pentosus YN-04, C is a length distribution graph of third-generation sequencing reads of Lactiplantibacillus plantarum YN-05, and D is a length distribution graph of third-generation sequencing reads of Lactiplantibacillus pentosus YN-04;

[0025] Figure 4 Figure 3 is a gene length distribution graph, wherein A is Lactiplantibacillus plantarum YN-04, and B is Lactiplantibacillus pentosus YN-05;

[0026] Figure 5 Figure 4 is a COG functional annotation graph, wherein A is Lactiplantibacillus plantarum YN-05, and B is Lactiplantibacillus pentosus YN-04;

[0027] Figure 6 Figure 5 is a GO functional annotation graph, wherein A is Lactiplantibacillus plantarum YN-05, and B is Lactiplantibacillus pentosus YN-04;

[0028] Figure 7 Figure 6 is a KEGG functional annotation graph of Lactiplantibacillus plantarum YN-05;

[0029] Figure 8 Figure 7 is an electronic nose and electronic tongue response value radar chart of fermented beancurd, wherein a is an electronic nose response value radar chart, b is an electronic tongue response value radar chart, A in the figure represents fermented beancurd fermented by Lactiplantibacillus pentosus YN-04, B in the figure represents fermented beancurd fermented by Lactiplantibacillus plantarum YN-05, C in the figure represents fermented beancurd fermented by a fermentation inoculant composed of Lactiplantibacillus pentosus YN-04 and Lactiplantibacillus plantarum YN-05, and D in the figure represents commercially available beancurd;

[0030] Figure 9 Figure 8 is a PCA analysis of metabolites of fermented beancurd, wherein a is a comparison between A and D, b is a comparison between B and D, and c is a comparison between C and D, A in the figure represents fermented beancurd fermented by Lactiplantibacillus pentosus YN-04, B in the figure represents fermented beancurd fermented by Lactiplantibacillus plantarum YN-05, C in the figure represents fermented beancurd fermented by a fermentation inoculant composed of Lactiplantibacillus pentosus YN-04 and Lactiplantibacillus plantarum YN-05, and D in the figure represents commercially available beancurd;

[0031] Figure 10Metabolites of fermented beancurd OPLS-DA analysis, wherein a is OPLS-DA score plot of A vs D, b is model scatter plot of A vs D, c is OPLS-DA score plot of B vs D, d is model scatter plot of B vs D, e is OPLS-DA score plot of C vs D, f is model scatter plot of C vs D, A in the figure represents beancurd fermented by Lactiplantibacillus pentosus YN-04, B in the figure represents beancurd fermented by Lactiplantibacillus plantarum YN-05, C in the figure represents beancurd fermented by the fermentation agent composed of Lactiplantibacillus pentosus YN-04 and Lactiplantibacillus plantarum YN-05, and D in the figure represents commercially available beancurd;

[0032] Figure 11 Volcano plot of differential metabolites of fermented beancurd, wherein A is the comparison between beancurd fermented by YN-04 strain and commercially available beancurd, B is the comparison between beancurd fermented by YN-05 strain and commercially available beancurd, and C is the comparison between beancurd fermented by the fermentation agent composed of Lactiplantibacillus pentosus YN-04 and Lactiplantibacillus plantarum YN-05 and commercially available beancurd;

[0033] Figure 12 KEGG bubble plot of differential metabolites of fermented beancurd, wherein A is the comparison between beancurd fermented by YN-04 strain and commercially available beancurd, B is the comparison between beancurd fermented by YN-05 strain and commercially available beancurd, and C is the comparison between beancurd fermented by the fermentation agent composed of Lactiplantibacillus pentosus YN-04 and Lactiplantibacillus plantarum YN-05 and commercially available beancurd;

[0034] Figure 13 Phylogenetic tree of Lactiplantibacillus plantarum YN-05;

[0035] Figure 14 Gene annotation statistics chart, wherein A is Lactiplantibacillus plantarum YN-05 and B is Lactiplantibacillus pentosus YN-04;

[0036] Figure 15 Genome circle chart of Lactiplantibacillus plantarum YN-05;

[0037] Figure 16 Morphology chart of Lactiplantibacillus pentosus YN-04 and Lactiplantibacillus plantarum YN-05, wherein A is Lactiplantibacillus pentosus YN-04 and B is Lactiplantibacillus plantarum YN-05, and the scale is 10 μm;

[0038] Figure 17 KEGG function annotation chart of Lactiplantibacillus pentosus YN-04;

[0039] Figure 18 Genome circle chart of Lactiplantibacillus pentosus YN-04;

[0040] Figure 19 A pie chart showing the classification and percentage of metabolites in a milk tofu sample.

[0041] Preservation Certificate

[0042] Lactobacillus pentosus YN-04, Latin name: Lactiplantibacillus pentosus YN-04, this strain is deposited at the China Center for Type Culture Collection, Wuhan University, Wuhan, China, on March 13, 2025, with accession number CCTCC NO: M 2025469.

[0043] Lactobacillus plantarum YN-05, Latin name Lactiplantibacillus plantarum YN-05, this strain is deposited at the China Center for Type Culture Collection, Wuhan University, Wuhan, China, on March 13, 2025, with accession number CCTCC NO: M 2025470. Detailed Implementation

[0044] Various exemplary embodiments of the present invention will now be described in detail. This detailed description should not be considered as a limitation of the present invention, but rather as a more detailed description of certain aspects, features, and embodiments of the present invention.

[0045] It should be understood that the terminology used in this invention is merely for describing particular embodiments and is not intended to limit the invention. Furthermore, with respect to numerical ranges in this invention, it should be understood that each intermediate value between the upper and lower limits of the range is also specifically disclosed. Every smaller range between any stated value or intermediate value within a stated range, and any other stated value or intermediate value within said range, is also included in this invention. The upper and lower limits of these smaller ranges may be independently included or excluded from the range.

[0046] Unless otherwise stated, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art. While only preferred methods and materials have been described herein, any methods and materials similar or equivalent to those described herein may be used in the implementation or testing of this invention. All references to this specification are incorporated by way of citation to disclose and describe methods and / or materials associated with those references. In the event of any conflict with any incorporated reference, the content of this specification shall prevail.

[0047] Various modifications and variations can be made to the specific embodiments described in this specification without departing from the scope or spirit of the invention, as will be apparent to those skilled in the art. Other embodiments derived from this specification will also be apparent to those skilled in the art. This specification and embodiments are merely exemplary.

[0048] The terms “include,” “including,” “have,” “contain,” etc., used in this article are all open-ended terms, meaning that they include but are not limited to.

[0049] The materials used in this invention are sourced from: grazing raw milk from herders in Abaga Banner, Bai Banner, and Zhenglan Banner of Xilingol League, Inner Mongolia; stall-fed raw milk from pastures in Abaga Banner, Bai Banner, and Zhenglan Banner of Xilingol League, Inner Mongolia; and buffers S, GA, GB, DV, BV, W1, and W2 containing RNase A from Illumina Corporation, USA.

[0050] The MRS broth culture medium (solid) used in this invention has the following formula: 10g peptone (animal source), 8g beef extract, 4g yeast extract, 20g glucose, 1mL Tween-80, 2g dipotassium hydrogen phosphate, 5g anhydrous sodium acetate, 2g triamine citrate, 0.05g manganese sulfate tetrahydrate, 0.2g magnesium sulfate heptahydrate, 15g agar powder, and 1L distilled water to adjust the initial pH to 6.20. The medium is then sterilized at 121℃ for 15min.

[0051] The MRS broth culture medium (liquid) used in this invention has the following formula: 10g peptone (animal source), 8g beef extract, 4g yeast extract, 20g glucose, 1mL Tween-80, 2g dipotassium hydrogen phosphate, 5g anhydrous sodium acetate, 2g triamine citrate, 0.05g manganese sulfate tetrahydrate, 0.2g magnesium sulfate heptahydrate, and 1L distilled water to bring the volume to a final volume. The initial pH is adjusted to 6.20, and the medium is sterilized at 121℃ for 15min.

[0052] This invention does not have any special limitations on the source of peptone (animal-derived), beef extract, yeast extract, glucose, Tween-80, dipotassium hydrogen phosphate, anhydrous sodium acetate, triamine citrate, manganese sulfate tetrahydrate, magnesium sulfate heptahydrate, and agar powder; conventional commercially available products well known to those skilled in the art can be used.

[0053] The skim milk culture medium used in this invention is: 100.0 g / L skim milk powder, heated to dissolve, dispensed, and autoclaved at 121°C for 15 min.

[0054] The cholesterol (TC) culture medium formula used in this invention is as follows: Accurately weigh 0.1g of cholesterol into a small beaker, add 1mL LTween-80 and 0.1g of sucrose ester, stir well, then add 5.0mL of glacial acetic acid, heat and stir until fully dissolved, sonicate for 15min, then add to MRS culture medium while stirring to make the cholesterol concentration 0.1mg / mL. Then add 0.2% sodium thioglycolate, adjust the pH to 6.0, and sterilize at 121℃ for 15min.

[0055] Example 1

[0056] I. Isolation and screening of Lactobacillus pentosolicus YN-04 and Lactobacillus plantarum YN-05.

[0057] 10g of fresh milk samples were collected from two groups (PXN group, grazing condition; H group, stall feeding condition; XN group, raw milk) under different feeding conditions, and mixed with 90mL of physiological saline. The mixture was shaken for 30min. The samples were serially diluted and plated for lactic acid bacteria isolation. After incubation at 37℃ for 48h, colony morphology was observed. Single colonies with distinct characteristics were then picked from MRS broth (solid) plates and streaked three times for purification. Gram staining and catalase assays were then performed. The 109 purified lactic acid bacteria strains were preserved using a 30% (v / v) glycerol solution and stored at -80℃ for later use.

[0058] II. Extraction of DNA from the bacterial strain.

[0059] First, prepare 1.0 × 10 9 CFU / mL bacterial culture medium was used, followed by centrifugation to remove the supernatant. The precipitate was then resuspended in buffer S containing RNase A. Next, lysozyme stock solution was added, mixed thoroughly, and allowed to stand for 5 min. EDTA was then added, mixed, and incubated on ice for 5 min. Buffer GA was then added, vortexed, and incubated at 65°C for 10 min. Buffer GB and pre-chilled buffer DV were then added, mixed, and centrifuged. The superphase was discarded, retaining the precipitate and lower phase, which were washed with buffer DV to remove the superphase. The filtrate was then centrifuged, followed by the addition of buffer BV, and centrifuged again. Finally, the precipitate was washed sequentially with buffer W1 and buffer W2, and DNA was eluted with deionized water. The extracted DNA was obtained by centrifugation.

[0060] III. PCR amplification of the strain's genome.

[0061] Table 1 PCR amplification reaction system

[0062] ;

[0063] The PCR amplification reaction system was prepared by adding the components shown in Table 1 to a 0.2 mL centrifuge tube. The tube was gently tapped to mix, and the liquid droplets on the tube wall were collected to the bottom after a brief centrifugation. The PCR reaction was then performed on a PCR instrument with the following parameters: pre-denaturation (95℃, 5 min), denaturation (95℃, 30 s), annealing (58℃, 30 s), extension (72℃, 90 s), and final extension (72℃, 7 min), for a total of 35 cycles.

[0064] After the reaction, 3 μL of PCR product was subjected to 1% agarose gel electrophoresis to confirm the PCR amplified fragments. Next, the PCR products were recovered using AxyPrep DNA gel electrophoresis, and the purified PCR products for each bacterial species were sent to an ABI 3730-XL sequencer for DNA sequencing. Simultaneously, sequences of highly homologous type strains were downloaded, and phylogenetic studies were constructed using the neighbor-joining (NJ) method with MEGA 7.0 software. The number and species of lactic acid bacteria were also counted and identified.

[0065] like Figure 1 The image shown is a phylogenetic tree of *Lactobacillus pentosaccharis* YN-04. Figure 13 The figure shown is a phylogenetic tree diagram of Lactobacillus plantarum YN-05.

[0066] The nucleotide sequences of the 16S rRNA of *Lactobacillus pentosus* YN-04 and *Lactobacillus plantarum* YN-05 are shown in SEQ ID NO.1 and SEQ ID NO.2, respectively.

[0067] IV. Screening of strains.

[0068] The isolated and purified lactic acid bacteria strains were cultured in MRS broth (liquid) for preliminary screening of target strains. The main objective of the screening was to identify strains with rapid acid production, good curdling effect, and rich flavor.

[0069] (1) Determination of the curdling characteristics and acid production capacity of the strain:

[0070] A. The isolated and purified strain was inoculated into a 100 g / L (sterilized) skim milk medium at an inoculation ratio of 3% (v / v), and then cultured at 25°C for 48 h. The time required for curdling and its state were observed.

[0071] B. Using the national food safety standard GB5009.239-2016, the acidity and pH of the curd during the curdling process were determined, and the acid production rate was calculated using the following formula:

[0072] △T = °T / t;

[0073] Where △T represents the average acid production rate, °T represents the fermentation acidity, and t represents the curdling time.

[0074] (2) The thermal coagulation properties of the strain were determined:

[0075] The purified strain was inoculated at a 3% (v / v) inoculation rate into 100 g / L (sterile) skim milk medium and cultured at 25°C for 48 h. After culturing, the curd sample was placed in a 65°C water bath and heated for 5 min, until the temperature reached approximately 85°C. The curd condition and flavor developed during heating were then observed and recorded. This process allowed strains with poor curd characteristics and undesirable flavors to be eliminated.

[0076] (3) Determination of the growth ability of the strain in skim milk:

[0077] The target strain obtained after initial screening was added to 100 g / L sterilized skim milk according to the addition amount of A, and fermented at a constant temperature of 25℃. Samples were taken at 0, 6, 12, 16, 18, 20 and 24 h, and its acidity and pH value were measured according to step B.

[0078] V. Colony morphology characteristics.

[0079] The *Lactobacillus pentosus* YN-04 and *Lactobacillus plantarum* YN-05 of the present invention form opaque, milky-white colonies with a diameter between 0.5 and 1 mm, round and flat or micro-papillary, with smooth edges, on MRS broth medium (solid).

[0080] Inside a clean bench, purified *Lactobacillus pentosaceus* YN-04 and *Lactobacillus plantarum* YN-05 bacterial slurry was picked up with an inoculation loop and smeared. After drying, Gram staining was performed. The morphological and staining characteristics of the strains were observed and recorded under a microscope. Figure 16 China A and Figure 16 As shown in Figure B, the strain is a Gram-positive bacterium observed under a microscope. The bacteria are rod-shaped and mostly appear as single, paired, chain, or clustered cells. On MRS broth medium (solid), the colony diameter and edges are regular, and the colony is creamy white and opaque.

[0081] The *Lactobacillus pentosolicus* YN-04 and *Lactobacillus plantarum* YN-05 of the present invention have the following biological characteristics: the bacterial cells are rod-shaped under a microscope, and their arrangement is also single, paired, chain-like or aggregated, but not in obvious chain-like structure, without flagella or cilia, and they do not move.

[0082] Example 2

[0083] The probiotic properties of Lactobacillus pentosolicus YN-04 and Lactobacillus plantarum YN-05.

[0084] I. Determination of acid and bile salt resistance:

[0085] After activating the strain for two generations, the strain was inoculated at a concentration of 3% (v / v) into MRS broth medium at pH 2.5 and MRS broth medium with a bovine bile salt concentration of 3%, and incubated at 37°C. Viable cell counts were performed at 0 and 3 hours to determine the survival rate of the strain. The formula for calculating the survival rate of the strain is as follows:

[0086] Survival rate (%) = N1 / N0 × 100%;

[0087] Where N1 represents the number of viable bacteria measured at 3h, CFU / mL; N0 represents the number of viable bacteria measured at 0h, CFU / mL.

[0088] The acid and bile salt resistance results of Lactobacillus pentosolicus YN-04 and Lactobacillus plantarum YN-05 are shown in Table 2 below.

[0089] Table 2. Acid and bile salt tolerance of *Lactobacillus pentosus* YN-04 and *Lactobacillus plantarum* YN-05

[0090] ;

[0091] II. Determination of the antioxidant capacity of the strain:

[0092] The MRS-activated bacterial culture was centrifuged at 5500 r / min for 15 min, washed twice with PBS to prepare a bacterial suspension (OD). 600 (Value 1.0) Reserved.

[0093] (1) DPPH free radical scavenging ability:

[0094] A 0.4 mmol / L DPPH solution was prepared using anhydrous ethanol. 2 mL of the DPPH solution was mixed with the bacterial suspension and incubated at room temperature in the dark for 30 min. Subsequently, the mixture was centrifuged at 6000 rpm for 10 min. The supernatant was collected after centrifugation, and its absorbance was measured at 517 nm. Three replicates were performed for each group. Finally, the DPPH free radical scavenging rate was calculated using the following formula:

[0095] DPPH free radical scavenging rate (%) = [1 - (A b -A c ) / A a ×100%;

[0096] Among them, A b The OD value is the result of mixing the bacterial suspension with the DPPH solution; A c The OD value is the result of mixing the bacterial suspension with anhydrous ethanol; A a This represents the OD value after mixing pure water with DPPH solution.

[0097] (2) OH scavenging ability:

[0098] Prepare 9 mmol / L FeSO4 solution, 9 mmol / L salicylic acid-ethanol solution, and 8.8 mmol / L H2O2 solution. Then, take 1 mL of the bacterial suspension and add 1 mL of each of the three solutions sequentially. Incubate at room temperature for 20 min and measure the absorbance at 510 nm. Perform three replicates per group. Calculate the hydroxyl radical scavenging rate using the following formula:

[0099] Hydroxyl radical scavenging rate (%) = [1-(B b -B a ) / B c ×100%;

[0100] Among them, B b B represents the OD value of the bacterial suspension and the mixed solution. a The OD value of pure water instead of H2O2 solution; B c This represents the OD value of the pure water and the mixed solution.

[0101] (3) Cholesterol capacity test:

[0102] The cholesterol degradation ability of the target strain was determined using the ferric phosphate-sulfur colorimetric method. The preparation process of the ferric phosphate-sulfur reagent is as follows: Accurately weigh 2.50 g of FeCl3·6H2O and dissolve it in 100 mL of phosphoric acid to obtain a FeCl3 solution. Then, take 88 mL of this solution, add concentrated sulfuric acid, and dilute to 100 mL to finally obtain the ferric phosphate-sulfur colorimetric reagent.

[0103] Plotting a cholesterol standard curve: Establishing a cholesterol standard curve (e.g.) Figure 2 As shown in the figure, its regression equation is y = 0.4629x + 0.0141, R0 2 =0.9948.

[0104] A cholesterol medium with a concentration of 0.1 mg / mL was prepared, and the cholesterol degradation rate of lactic acid bacteria was determined using the phosphorus-sulfur-ferric colorimetric method. Activated VN-04 and YN-05 strains were inoculated at 3% onto the cholesterol medium, with an uninoculated control group. After incubation at 37°C for 24 h, the culture medium was mixed with anhydrous ethanol and centrifuged. The supernatant was then added to the phosphorus-sulfur-ferric colorimetric reagent, cooled to room temperature, and the absorbance was measured at 550 nm. Three replicates were performed for each group.

[0105] The formula for calculating cholesterol degradation rate is as follows:

[0106] Cholesterol degradation rate (%) = (CA) / C × 100%;

[0107] Where C represents the cholesterol concentration (mg / mL) of the blank control group; A represents the cholesterol concentration (mg / mL) of the sample.

[0108] The antioxidant capacity of Lactobacillus pentosolicus YN-04 and Lactobacillus plantarum YN-05 includes DPPH free radical scavenging rate, hydroxyl free radical scavenging capacity and cholesterol degradation rate, as shown in Table 3 below.

[0109] Table 3. DPPH radical scavenging rate, hydroxyl radical scavenging capacity, and cholesterol degradation rate of the strains

[0110] ;

[0111] III. Antibacterial ability test:

[0112] The double-layer agar diffusion method was used, with *Escherichia coli*, *Staphylococcus aureus*, and *Salmonella typhimurium* selected as control bacteria. The Oxford cup method was then used to measure the antibacterial activity of the two target strains. First, sterilized petri dishes were poured into 10 mL of nutrient agar, allowed to cool slightly, and then placed in three Oxford cups. Next, 1 mL of indicator bacterial solution (concentration 10) was added. 7 After mixing (CFU / mL) with 10 mL of nutrient agar, allow it to solidify. Then, remove the Oxford cup and add 0.2 mL of lactic acid bacteria fermentation broth into the well. Finally, incubate at 37°C for 24 hours and measure the diameter of the inhibition zone.

[0113] The inhibitory effects of Lactobacillus pentosolicus YN-04 and Lactobacillus plantarum YN-05 on Escherichia coli, Staphylococcus aureus and Salmonella typhimurium are shown in Table 4 below.

[0114] Table 4. Antibacterial activity of the strains against Escherichia coli, Staphylococcus aureus, and Salmonella typhimurium.

[0115] ;

[0116] Example 3

[0117] I. Whole genome sequencing of the strain.

[0118] Purification and detection of strain DNA: The sequencing strain was cultured at 37°C for 24 hours, 1 mL of bacterial cells were collected, the bacterial cells were stored on dry ice, and sent to Meiji Biotechnology Co., Ltd. for whole genome sequencing.

[0119] Illumina library preparation and Denova sequencing: Whole genome sequencing was performed using both Illumina HiSeq and PacBio sequencing platforms. First, the extracted DNA was fragmented to generate a DNA fragment of approximately 10kb. Through a series of steps, including end-completion, 3' A-tailing, SMRT sequencing adapters at both ends, library preparation, paired-end sequencing, and data analysis, sequencing data reads were finally obtained.

[0120] Data quality control and sequence assembly: The raw sequencing data underwent quality trimming, removing low-quality reads, reads with a high proportion of nitrogen (N), and reads that were too short, resulting in high-quality processed data. Unicycler software and the PacBio database were used for data assembly. Finally, Illumina data were used to optimize and correct the assembly results, yielding the final data.

[0121] Gene prediction and annotation: Glimmer software was used for coding gene prediction, GeneMarkS (4.3) software was used for plasmid gene prediction, and tRNAs, rRNAs, and sRNAs were predicted and annotated using tRNAscan-SE (2.0.12), Infernal (1.1.4), and the Rfam database, respectively. Scattered repeats and tandem repeats were predicted using Repeatmasker and Tandem Repeats Finder, respectively. Genome islands, prophages, and CRISPR-Cases were predicted using IslandViewer (1.2), Phage Finder, and Minced (0.2.0), respectively.

[0122] Protein function annotation of CDS was performed using tools such as BLAST, Diamond, and HMMER, combined with databases including NR, GO, COG, and KEGG. Secondary metabolite gene clusters were identified and analyzed using antiSMASH. Simultaneously, Circos was used to construct genome circles for YN-04 and YN-05, providing bioinformatics support for subsequent research.

[0123] II. Basic genome-wide characteristics of Lactobacillus pentosolicus YN-04 and Lactobacillus plantarum YN-05.

[0124] Following DNA extraction and purification, sequencing libraries were constructed from *Lactobacillus pentosus* YN-04 and *Lactobacillus plantarum* YN-05. Subsequently, their complete genomes were sequenced using a combination of second-generation Illumina sequencing technology and third-generation PacBio sequencing technology.

[0125] After filtering, the raw data from third-generation sequencing of *Lactobacillus plantarum* YN-05 and *Lactobacillus pentosaccharide* YN-04 yielded 222.39 Mb and 109.79 Mb bases, respectively, totaling 26038 and 13332 reads. The longest read lengths were 38504 bp and 19495 bp, and the average read lengths were 8540.99 bp and 8234.85 bp, respectively. The Q30 accuracy was 97.28% and 95.62%, and the Q20 accuracy was 99.19% and 98.64%, respectively. Based on the Illumina requirements of Q20 > 80% and Q30 > 70% (Q20: the percentage of bases with a base recognition accuracy greater than 99%; Q30: the percentage of bases with a base recognition accuracy greater than 99.9%), the raw data from the sequencing of *Lactobacillus pentosaccharide* YN-04 and *Lactobacillus plantarum* YN-05 are accurate.

[0126] When using Illumina sequencing technology, a massive amount of read data is generated. This data is so large that it is impossible to show the specific quality status of each read in detail. Therefore, it is necessary to use statistical analysis methods to comprehensively evaluate the library construction quality and sequencing quality of the samples from a macroscopic perspective, including base composition distribution maps and base quality distribution statistics.

[0127] like Figure 3 China A and Figure 3 Figure B shows the base composition distribution of pre-sequencing reads for two strains, *Lactobacillus plantarum* YN-05 and *Lactobacillus pentosus* YN-04, respectively. The horizontal axis represents the base position of the reads, arranged sequentially from the 5' to the 3' end, while the vertical axis reflects the percentage of A, C, G, T, and N bases at each sequencing position. Different bases are represented by different colors. At the beginning of the sequence, the proportions of A, C, G, and T bases fluctuate due to the influence of sequencing primer adapters; however, as the sequence continues, the base proportions tend to stabilize. The figure shows that the lower proportion of N bases indicates fewer unknown bases in the sample, suggesting higher sequencing data quality and less influence from systematic AT bias. These results indicate that the sequencing data after quality control has good accuracy, a relatively stable base composition, and that biases during the sequencing process have been effectively controlled.

[0128] like Figure 3 C and Figure 3Figure D shows the clean read length distribution of third-generation sequencing for two strains, *Lactobacillus plantarum* YN-05 and *Lactobacillus pentosus* YN-04. The horizontal axis represents the length of the sequencing reads, and the vertical axis represents the number of reads of different lengths. The black curve reflects the change in the number of reads as the length increases; the green curve represents the change in the cumulative total number of bases as the length increases, although the green area has no practical significance. As can be seen from the figure, the third-generation sequencing data of strains YN-04 and YN-05 have relatively long reads and no uncertain bases. Therefore, by splicing and assembling the third-generation sequencing data and combining it with the second-generation data, a complete genome sequence was finally obtained.

[0129] In summary, the genome sequencing and assembly of *Lactobacillus pentosus* YN-04 and *Lactobacillus plantarum* YN-05 were satisfactory, and the sequencing depth met the expected requirements. Further bioinformatics analysis can be carried out next.

[0130] The basic genomic information of two strains, Lactobacillus pentosolicus YN-04 and Lactobacillus plantarum YN-05, is shown in Table 5 below.

[0131] Table 5. Basic genomic characteristics of *Lactobacillus pentosus* YN-04 and *Lactobacillus plantarum* YN-05.

[0132] ;

[0133] As shown in Table 5, a total of 2558 genes were identified in *Lactobacillus plantarum* strain YN-05, of which 67 tRNA sequences and 19 rRNA sequences were successfully predicted. *Lactobacillus pentosus* strain YN-04, on the other hand, identified 3156 genes, including 68 tRNA sequences and 16 rRNA sequences. Figure 4 The YN-04 gene length distribution diagram shown in Figure A and Figure 4 As shown in Figure B, the gene length distribution of YN-05 strains shows that most genes are concentrated between 200 and 1000 bp in length, but there are also a certain proportion of long genes (greater than 1000 bp). In strain YN-04, 984 genes are longer than 1000 bp, accounting for 31.18% of all coding genes. In strain YN-05, 737 genes are longer than 1000 bp, accounting for 31.1828.81% of all coding genes. These results indicate that although most genes are of moderate length, some longer genes still exist in both strains, demonstrating the diversity and complexity of their genome structure.

[0134] III. tRNA gene prediction.

[0135] Lactobacillus pentosolicus YN-04 and Lactobacillus plantarum YN-05 contained 68 and 67 tRNAs, respectively. Table 6 below shows the statistical results of tRNA prediction.

[0136] Table 6 Statistical Table of tRNA Prediction Results

[0137] ;

[0138] The tandem repeat sequences of the two strains were predicted using the Tandem Repeats Finder software. The results showed that the genomes of YN-05 and YN-04 contained 36 and 71 repeat sequences, respectively, with total lengths of 9420 bp and 24404 bp, accounting for 0.44% and 0.88% of the genome, respectively. These data provide crucial clues for analyzing the genomic structure and DNA characteristics of the two strains and are of significant importance in gene evolution, expression regulation, and disease resistance.

[0139] Scattered repetitive sequences (transposon elements) include DNA transposons and retrotransposons (such as LTR, LINE, SINE, etc.). Repeatmasker software was used to predict the genomes of the two strains, and the results showed that YN-05 and YN-04 contained 4 and 2 DNA transposons, respectively, and the retrotransposons were SINE (11 and 10), LINE (11 and 9), and LTR (0 and 2), respectively.

[0140] Example 4

[0141] I. COG Function Comments.

[0142] like Figure 5 China A and Figure 5 As shown in Figure B, strains YN-05 and YN-04, respectively, have 2129 and 2748 genes with completed protein annotations, respectively. According to functional classification, the most numerous genes belong to metabolic functions (including classes C, E, F, G, H, I, P, and Q), with 855 in YN-05 and 1094 in YN-04; followed by genes related to cellular processes and signals (including classes D, M, N, O, T, U, V, and W), with 506 and 695 respectively; genes related to information storage and processing (including classes D, M, N, O, T, U, V, and W), with 487 and 616 respectively; and genes with unclear functions (class S), with 281 and 343 respectively.

[0143] The functions with the most annotations in strain YN-05 are: translation, ribosome structure and biosynthesis (J, 202), amino acid transport and metabolism (E, 183), carbohydrate transport and metabolism (G, 178), transcription (K, 171), routine function prediction (R, 166), and cell wall / cell membrane / extracellular membrane biosynthesis (M, 143). The functions with the most annotations in strain YN-04 are: transcription (K, 262), carbohydrate transport and metabolism (G, 244), amino acid transport and metabolism (E, 255), translation, ribosome structure and biosynthesis (J, 222), routine function prediction (R, 219), and cell wall / cell membrane / extracellular membrane biosynthesis (M, 165).

[0144] II. GO function comments.

[0145] According to comparisons with GO annotation databases, strains YN-05 and YN-04 have 926 and 1147 genes related to cell composition (CC), 1490 and 1908 genes related to molecular function (MF), and 952 and 1164 genes related to biological processes (BP), respectively. Strains YN-05 and YN-04 have 1811 and 2314 genes annotated in these three GO classifications, respectively, accounting for 70.8% and 73.32% of the total number of coding genes.

[0146] like Figure 6 China A (YN-05) and Figure 6 As shown in Figure B (YN-04), comparisons with the GO database reveal that in strain YN-05, the most frequently annotated functional components in cell composition are the cell membrane, cytoplasm, plasma membrane, and ribosomes, suggesting these encoding genes may possess transmembrane transport capabilities. In molecular function annotations, ATP binding, DNA binding, and hydrolytic enzyme activity are frequently noted. In biological processes, phosphorylation and translation are frequently annotated. In strain YN-04, the most frequently annotated functional components in cell composition are membrane components and the cytoplasm. In molecular function annotations, attachment activity is frequently noted, primarily because the interactions between biomolecules are crucial in biological reactions, with attachment being a significant manifestation. Specifically, the ATP binding mechanism is a prime example. ATP-binding genes specifically attach to ATP molecules, thereby acting on their phosphate groups to catalyze ATP hydrolysis, releasing energy for other biological processes. This process is not only a core link in energy conversion but also fundamental to the smooth operation of numerous intracellular metabolic activities. In biological processes, phosphorylation and proteolysis are frequently annotated.

[0147] III. KEGG Function Comments.

[0148] The KEGG database provides functional annotations for the whole genomes of *Lactobacillus pentosaceus* YN-04 and *Lactobacillus plantarum* YN-05. KEGG is a specialized database for comparing and analyzing bacterial genes and metabolic pathways at the molecular level, enabling exploration of bacteria at the molecular level and studying the relationships between functional genes and biomolecules in various biological pathways. The KEGG database categorizes microbial biological pathways into six major classes: metabolism, genetic information processing, environmental information processing, cellular processes, organismal systems, and human diseases. For example... Figure 7 (YN-05) and Figure 17 As shown in (YN-04), KEGG annotations revealed 1980 and 2281 functional genes for *Lactobacillus pentosus* YN-04 and *Lactobacillus plantarum* YN-05, respectively, mainly focusing on metabolism, especially carbohydrate (PXN 179; HXN 196) and amino acid and nucleotide metabolism (PXN 147; HXN 135), indicating that these two strains have strong metabolic capabilities.

[0149] IV. CAZy Function Comments.

[0150] The Carbohydrate Active Enzyme Database (CAZy) is used to annotate carbohydrate-related enzymes, categorized into six classes: glycoside hydrolases, glycosyltransferases, polysaccharide lyases, carbohydrate esterases, carbohydrate-binding modules, and helper oxidoreductases. According to the CAZy database, *Lactobacillus pentosus* YN-04 and *Lactobacillus plantarum* YN-05 have annotated 73 and 96 carbohydrate active enzyme genes, respectively, belonging to 4 (YN-05) and 5 (YN-04) protein families. YN-05 and YN-04 contain 2 and 11 helper oxidoreductase genes, 13 and 18 carbohydrate esterase genes, 37 and 39 glycoside hydrolases, and 21 and 27 glycosyltransferase genes, respectively. YN-04 also contains one carbohydrate-binding module gene.

[0151] And by Figure 14 China A and Figure 14 As shown in section B, in *Lactobacillus plantarum* YN-05 and *Lactobacillus pentosus* YN-04, auxiliary oxidoreductases accounted for 2.74% and 11.76% of the total number of carbohydrate active enzyme genes, respectively; carbohydrate esterase genes accounted for 17.81% and 18.75%, respectively; glycoside hydrolase genes accounted for 50.68% and 40.63%, respectively; and glycosyltransferase genes accounted for 28.78% and 28.13%, respectively. In contrast, the proportion of carbohydrate binding module genes in strain YN-04 was 1.04%.

[0152] V. Genome Circle Map.

[0153] like Figure 15 and Figure 18The diagram shows the genome loops of strains YN-05 and YN-04. This loop is clearly layered, revealing various key characteristics of the genome from the outside in. The first and fourth loops label the coding sequences (CDS) on the positive and negative strands, respectively. Different colors are carefully used to distinguish COG functional categories, allowing readers to easily identify the distribution of various functional genes. The second and third loops show the distribution of CDS, tRNA, and rRNA on the positive and negative strands, providing strong support for a deeper understanding of the transcription and translation processes of the genome. The fifth loop focuses on GC content. The outward-protruding portion directly reflects that the GC content in this region is higher than the average level of the whole genome, and the height of the peak directly corresponds to the size of the difference from the average GC content. Conversely, the inward-concave portion indicates that the GC content in this region is lower than the average of the whole genome; similarly, the significance of the peak directly reflects the degree of difference from the average GC content. This layered analysis helps to reveal the base composition preferences in the genome and their potential functional significance. The sixth ring displays the GC-Skew value, visually illustrating the imbalanced distribution of G and C bases in the genome through a stark contrast between green (skew+) and purple (skew-). Skew+ indicates a higher G content than C, while skew- indicates the opposite, meaning a lower G content than C. This value is calculated as (GC) / (G+C). It not only helps us determine the leading and lagging strands (generally, leading strand GC skew > 0, lagging strand GC skew < 0), but also plays a crucial role in determining the origin of replication (minimum cumulative offset) and the endpoint (maximum cumulative offset), especially in the analysis of circular genomes. The innermost ring clearly indicates the size of the genome.

[0154] VI. Analysis of genes related to flavor substance metabolism.

[0155] The unique flavor of fermented dairy products is formed through complex biochemical reactions involving lactic acid bacteria (LAB) and enzymes. LAB metabolizes lactose to produce lactic acid, which in turn alters the taste of dairy products. Flavor formation primarily involves carbohydrate conversion, fat breakdown, and amino acid metabolism. Key enzymes in carbohydrate conversion include citrate lyase, oxaloacetate decarboxylase, and lactate dehydrogenase; key enzymes in amino acid metabolism include peptidases, transaminases, hydroxy acid dehydrogenases, and phosphotransferases.

[0156] Lactic acid bacteria (LAB) are typical auxotrophic microorganisms, relying primarily on external acquisition of essential amino acids for growth due to their limited amino acid synthesis capabilities. These amino acids are obtained mainly through two pathways: hydrolysis of casein in milk and direct absorption of free amino acids from the environment. Amino acids are not only essential nutrients for LAB growth but also important precursors for flavor compounds generated during their metabolism. According to the whole-genome sequencing analysis of YN-05 and YN-04 as recorded in Tables 7 and 8, YN-05 has 183 genes involved in amino acid synthesis and 178 genes involved in carbohydrate metabolism. Genes related to flavor synthesis include gene0444 (L-lactate dehydrogenase), gene1294 (L-lactate dehydrogenase), and gene0455 (β-galactosidase). YN-04 has 225 genes involved in amino acid synthesis and 244 genes involved in carbohydrate metabolism. Genes involved in flavor synthesis include gene0205 (acetate kinase), gene0297 (acetate kinase), and gene0314 (ethanol dehydrogenase).

[0157] Table 7 Genes related to flavor production metabolism in YN-05

[0158] ;

[0159] Continued from Table 7

[0160] ;

[0161] Continued from Table 7

[0162] ;

[0163] Continued from Table 7

[0164] ;

[0165] Continued from Table 7

[0166] ;

[0167] Continued from Table 7

[0168] ;

[0169] Continued from Table 7

[0170] ;

[0171] Continued from Table 7

[0172] ;

[0173] Continued from Table 7

[0174] ;

[0175] Table 8 Genes related to flavor production metabolism in YN-04

[0176] ;

[0177] Continued from Table 8

[0178] ;

[0179] Continued from Table 8

[0180] ;

[0181] Continued from Table 8

[0182] ;

[0183] Continued from Table 8

[0184] ;

[0185] Continued from Table 8

[0186] ;

[0187] Continued from Table 8

[0188] ;

[0189] Continued from Table 8

[0190] ;

[0191] Continued from Table 8

[0192] ;

[0193] Continued from Table 8

[0194] ;

[0195] Continued from Table 8

[0196] ;

[0197] Continued from Table 8

[0198] ;

[0199] While acetate kinase, phosphoacetyltransferase, and alcohol dehydrogenase were not found in YN-05, it possessed abundant aminopeptidase and amino acid synthases, enabling the production of various amino acids. YN-04, on the other hand, possessed these related enzymes, while YN-05 contained L-lactate dehydrogenase and malate / lactate dehydrogenase, enzymes absent in YN-04. The related enzymes of the two strains exhibited a complementary relationship. Both strains possessed genes involved in carbohydrate, amino acid, and lipid metabolism. Therefore, in-depth research into the flavor formation mechanisms and metabolic pathways of these two strains will help predict their flavor synthesis potential and provide theoretical support for their industrial applications.

[0200] VII. Functional gene characteristics of Lactobacillus pentosolicus YN-04 and Lactobacillus plantarum YN-05.

[0201] Tolerance-related functional gene annotation was performed on *Lactobacillus pentosus* YN-04 and *Lactobacillus plantarum* YN-05, and 9 protease-related genes (htpX, ftsH, clpC, clpE, gluP, ysxB, hslU, clpX, rseP) and 10 protease-related genes (gluP, ftsH, clpX, clpE, ysxB, clpC, clpP, rseP, dam, degP) were annotated, respectively.

[0202] Lactobacillus pentosaccharis YN-04 and Lactobacillus plantarum YN-05 were compared with the VFDB and VirulenceFinder databases, with sequence similarity >90% and sequence coverage >60% as screening criteria. The results showed that no potentially virulence-related genes were detected in the ProSci-246 genome. Genes related to proteases were detected and annotated in the genomes of Lactobacillus pentosaccharis YN-04 and Lactobacillus plantarum YN-05; specific genes are shown in Table 9 below.

[0203] Table 9. Protease-related genes in the genomes of *Lactobacillus pentosaccharide* YN-04 and *Lactobacillus plantarum* YN-05

[0204] ;

[0205] Example 5

[0206] I. Preparation of milk tofu auxiliary fermentation agent using Pentosacchariformis Lactobacillus YN-04 and Lactobacillus plantarum YN-05.

[0207] Raw milk → sterilization → cooling → adding starter culture → heating to remove whey → kneading into a ball → cutting → pressing → compression molding → finished milk tofu.

[0208] Key points: *Lactobacillus pentosolicus* YN-04 and *Lactobacillus plantarum* YN-05 were mixed in a mass ratio of (1:0, 0:1, 1:1) to form a bacterial suspension. This suspension was then inoculated at a 3% inoculation rate (v / v) into a sterilized raw milk sample (raw milk was placed in an 80℃ water bath, stirred for 15 minutes, and then cooled). The sample was placed in a 25℃ constant temperature incubator under simulated natural fermentation conditions. Fermentation reached its endpoint when the pH reached approximately 4.6, the acidity reached 65°T, and a small amount of whey separated. The curd was then poured into a beaker and placed in a 65℃ water bath. Continuous stirring was maintained to ensure even heating. Whey was continuously drained until the curd became smaller, shrunken, and elastic. The curd was then kneaded into a ball, cut, and placed in a mold to compress into a milk tofu sample.

[0209] Milk tofu fermented with *Lactobacillus pentosus* YN-04 (1:0) was classified as Group A; milk tofu fermented with *Lactobacillus plantarum* YN-05 (1:0) was classified as Group B; milk tofu fermented with *Lactobacillus pentosus* YN-04 and *Lactobacillus plantarum* YN-05 strains (1:1) was classified as Group C; and commercially available milk tofu was classified as Group D. The physicochemical properties of the prepared milk tofu are shown in Table 10, and the sensory characteristics are as follows: Figure 8 a and Figure 8 As shown in Figure b, the moisture content of group C was significantly lower than that of the other three groups (p < 0.05). Low moisture content helps in the preservation and maintenance of milk tofu and effectively inhibits the growth and reproduction of harmful microorganisms. Meanwhile, the ash and protein content of group C were significantly higher than those of the other three groups (p < 0.05). Ash content is an important indicator for evaluating the nutritional value of food, and its level reflects the overall value of the food. The protein components of milk tofu play a crucial role in the coagulation and dehydration shrinkage of milk, and changes in the relative content of specific protein components significantly affect the quality of traditional cheese. Compared to group D, there was no significant difference in fat content among the three fermentation groups.

[0210] Table 10 Physicochemical Indicators of Milk Tofu Prepared in Different Groups

[0211] ;

[0212] II. Qualitative and quantitative analysis of milk tofu metabolites based on GC-MS technology.

[0213] (1) Extraction of metabolites.

[0214] Weigh 50 mg of sample and add 500 μL of pre-cooled extraction buffer (methanol:chloroform, volume ratio 3:1, containing L-2-chlorophenylalanine). Vortex for 30 s, add steel beads, grind using a 40 Hz grinder for 4 min, and then sonicate in an ice-water bath for 5 min (repeat 3 times). Centrifuge the sample at 12000 rpm for 15 min at 4 °C, transfer 100 μL of supernatant to a 1.5 mL EP tube and dry. Combine 40 μL of each sample into a QC sample. Then, add 40 μL of methoxyamine salt reagent and incubate at 80 °C for 30 min, followed by 60 μL of BSTFA and incubate at 70 °C for 1.5 h. Finally, cool to room temperature, add 5 μL of FAMEs, and perform randomized analysis.

[0215] (2) GC-MS detection conditions.

[0216] A SHIMADZUGC-2030 gas chromatograph coupled with a QP2020 NX mass spectrometer was used, employing a DB-5MS column (30m × 250μm × 0.25μm). Samples were injected in 1μL volume in splitless mode. Helium was used as the carrier gas, with a purge flow rate of 3 mL / min at the injection port and a gas flow rate of 1 mL / min within the column. The column temperature program was as follows: initial temperature 50℃, hold for 1 min, then ramp to 310℃ at a rate of 8℃ / min and hold at 310℃ for 11.5 min. The injection port, transfer line, and ion source temperatures were 280℃, 280℃, and 200℃, respectively. The ionization voltage was -70 eV. Mass spectrometry data were acquired in full scan mode, with a mass range (m / z) of 50–500, an acquisition rate of 12.5 spectra per second, and a solvent delay time of 7.2 min.

[0217] (3) Data processing.

[0218] The mass spectrometry data were analyzed using ChromaTOF software (V 4.3x, LECO), including peak extraction, baseline correction, deconvolution, peak integration, and peak alignment. For qualitative analysis of the substances, the LECO-Fiehn Rtx5 database was used, including mass spectrometry matching and retention time index matching. The retention time indices are shown in Table 11 below.

[0219] Table 11 Retention Time Index Table

[0220] ;

[0221] Four types of milk tofu samples were analyzed using GC-MS technology, such as... Figure 19As shown, a total of 645 metabolites in 12 categories were identified. The four main groups of substances in milk tofu included organic acids and their derivatives, organic oxygen compounds, lipids and lipid molecules, and organic heterocyclic compounds, accounting for 72.56% of the total. Organic acids enhance the flavor of milk tofu and impart probiotic functions, improving its texture and positively impacting gut health. Lipids play a crucial role in the production of milk tofu, significantly affecting its flavor and preservation. The flavor of milk tofu largely depends on fat breakdown and its interactions with other components. Organic heterocyclic compounds are highly regarded in the flavoring industry; many possess unique aromas and have extremely low olfactory thresholds. Therefore, differences in the types and amounts of these compounds in milk tofu may be key factors contributing to variations in its flavor.

[0222] PCA analysis of metabolites from fermented milk tofu strains as follows: Figure 9 a, Figure 9 b and Figure 9 As shown in Figure c, the samples are located in different quadrants, and the clustering of samples between groups indicates significant differences. Principal component analysis can be used to differentiate samples of fermented milk tofu from different bacterial strain combinations. The OPLS-DA model score plots and validation results for each group are shown below. Figure 10 In the middle, 'a' represents the scatter plot of the OPLS-DA model for groups A and D. Figure 10 In the middle, b is a scatter plot of the OPLS-DA permutation test for groups A and D. Figure 10 In the middle, c represents the scatter plot of the OPLS-DA model for groups B and D. Figure 10 In the middle, d represents the scatter plot of the OPLS-DA permutation test between groups B and D. Figure 10 In the middle, e represents the scatter plot obtained from the OPLS-DA model for groups C and D. Figure 10 f is a scatter plot of the OPLS-DA permutation test for groups C and D. Figure 10 As shown in section a, the milk tofu fermented by *Lactobacillus pentosus* YN-04 (group A) and the commercially available milk tofu sample (group D) are located in different quadrants and are far apart from each other, indicating that there are significant differences in the metabolites between the milk tofu fermented by *Lactobacillus pentosus* YN-04 and the commercially available milk tofu. Similarly, from... Figure 10 c and Figure 10 As can be seen from Figure e, the milk tofu fermented with *Lactobacillus plantarum* YN-05 (represented by group B) has significantly different metabolites from commercially available milk tofu (represented by group D). Similarly, the milk tofu fermented with the fermentation agent composed of *Lactobacillus plantarum* YN-04 and *Lactobacillus plantarum* YN-05 (represented by group C) also shows significant differences from the metabolites of commercially available milk tofu (represented by group D). Figure 10 b, Figure 10 d and Figure 11As shown in f, the OPLS-DA substitution points are all located on the left side, and the intersection of its regression line with the vertical axis is significantly lower than the R² and Q² values ​​of the original model. This indicates that the original model is not overfitted and the results are significantly effective. This proves that the metabolites of the milk tofu prepared by groups A, B, and C are significantly different from those of commercially available milk tofu.

[0223] Depend on Figure 11 As shown in Figure A, a total of 148 differentially expressed metabolites were screened in groups A and D, mainly including organic acids and their derivatives, organic oxygen compounds, lipids and lipid-like molecules, organic nitrogen compounds, benzene and substituted derivatives, as well as nucleosides, nucleotides and their analogues. Among them, 38 differentially expressed metabolites, such as glycolic acid, L-valine, D-glutamate and citric acid, were upregulated, while 110 differentially expressed metabolites, such as maleimide, L-alanine, lactamide, and caprylic acid, were downregulated.

[0224] Depend on Figure 11 As shown in Figure B, a total of 177 differentially expressed metabolites were screened in groups B and D, mainly categorized into lipids and lipid-like molecules, organic acids and their derivatives, organic oxygen compounds, nucleosides, nucleotides and their analogues, phenylpropane and polyketides, and organic heterocyclic compounds. Among them, 75 differentially expressed metabolites, including L-homoserine, L-alanine, glycerophosphates, and myristic acid, were upregulated, while 102 differentially expressed metabolites, including pyruvate, lactamide, L-malate, and 3-aminoglutaric acid, were downregulated.

[0225] Depend on Figure 12 As shown in Figure C, a total of 148 differentially expressed metabolites were screened from groups C and D, mainly categorized into lipids and lipid-like molecules, organic acids and their derivatives, organic oxygen compounds, phenylpropane and polyketide compounds, and organic heterocyclic compounds. Among them, 144 differentially expressed metabolites, including L-alanine, glycine, L-homoserine, and lauric acid, were upregulated, while 35 differentially expressed metabolites, including sorbitol, lactobionic acid, hippuric acid, and 4-hydroxyquinoline, were downregulated.

[0226] like Figure 12 China A Figure 12 China B and Figure 12 As shown in Figure C, this is a bubble diagram illustrating the differences in metabolic pathways between the fermentation group and the commercially available group. Figure 12 As shown in Figure A, compared with Group D, Group A has six significant metabolic pathways, including arginine and proline metabolism, glyoxylic acid and dicarboxylic acid metabolism, pyrimidine metabolism, glycerol lipase metabolism, linoleic acid metabolism, and the biosynthesis of valine, leucine, and isoleucine (pathway p<0.05). Figure 12 In Figure B, the metabolic pathway enrichment map for groups B and D shows that both groups exhibit six significant metabolic pathways (p < 0.05): valine, leucine, and isoleucine biosynthesis; arginine and proline metabolism; glycerol lipase metabolism; pyrimidine metabolism; glycine, serine, and threonine metabolism; and linoleic acid metabolism. ​As shown in Figure C, compared with Group D, Group C had nine significant metabolic pathways, including arginine and proline metabolism, glycine, serine and threonine metabolism, valine, leucine and isoleucine biosynthesis, pantothenic acid and coenzyme A biosynthesis, glycerol metabolism, aminoacyl-tRNA biosynthesis, sulfur metabolism, linoleic acid metabolism, and arachidonic acid metabolism (pathway p<0.05).

[0227] In metabolite and KEGG enrichment analysis, significantly enriched metabolic pathways were found in groups C and D, with amino acid-related pathways showing a marked increase compared to single-strain fermentation. Among the nine differentially metabolized pathways, amino acid-related pathways were predominant, involving common differentially metabolites such as proline, creatine, L-threonine, glycine, pyruvate, L-valine, L-isoleucine, and succinic acid. Proline and glycine, with their slightly sweet taste, can mitigate any bitterness or other irritating flavors that may be present in cheese curd, resulting in a more balanced and mellow flavor. Creatine is a naturally occurring compound primarily found in animal-derived foods and is also a widely used functional dietary supplement in sports nutrition and health. It has several important biological functions in the body, including enhancing energy metabolism, promoting muscle cell growth and repair, and exerting antioxidant effects. L-valine is degraded through Strecker to produce isobutyraldehyde, a volatile compound with nutty and malty flavors that significantly enhances the flavor of cheese. L-Isoleucine is degraded by Strecker to produce 2-methylbutyraldehyde, a volatile compound with nutty and caramel flavors that enhances the flavor complexity of cheese.

[0228] III. Changes in the quality of milk tofu during storage.

[0229] The above-mentioned fermentation agent was used to prepare milk tofu as the experimental group, and traditional milk tofu (purchased from Changhong Dairy Products Factory, Zhenglan Banner, Xilingol League, Inner Mongolia) was selected as the control group. The changes in physicochemical indicators and total bacterial count of the two groups of milk tofu during storage (0, 15, 30, and 45 days) were investigated, as shown in Tables 12 and 13.

[0230] Milk tofu was prepared using the above fermentation agent: raw milk was sterilized (the raw milk was placed in an 80℃ water bath, stirred for 15 minutes, and then cooled), and the fermentation agent (the inoculation amount of the fermentation agent was calculated as 3% of the volume fraction of the sterilized raw milk) was inoculated into the sterilized raw milk. Fermentation was carried out at 25℃ until the pH reached 4.6. The whey was removed by heating in a water bath at 65℃, the mixture was kneaded into a ball, cut, and compressed into shape to obtain milk tofu.

[0231] Table 12 Changes in physicochemical properties of milk tofu during storage

[0232] ;

[0233] Note: Different letters in the same column indicate significant differences (P < 0.05); the same letters in the same column indicate no significant differences (P > 0.05).

[0234] Table 13 Changes in total bacterial count of milk tofu during storage period

[0235] ;

[0236] During storage, traditional milk tofu exhibits significant changes in its basic physicochemical and microbiological indicators. Table 12 shows that with prolonged storage, both fat and protein in the control group's traditional milk tofu showed significant degradation, resulting in substantial quality loss. Compared to traditional milk tofu, the experimental group's milk tofu showed trace fat degradation in the early stages of storage, which stabilized after 15 days, indicating that severe fat oxidation did not occur during the overall storage process. Simultaneously, around 30 days of storage, the experimental group's milk tofu showed some protein degradation, but the protein retention was significantly better than the control group, suggesting that the fermenting agent is beneficial for maintaining the storage quality of the milk tofu.

[0237] Table 13 shows that, to simulate the refrigerated storage conditions during the sales process, the study set 10℃ as the storage temperature for milk tofu. Under this condition, the total bacterial count of traditional milk tofu increased from the initial 3.42 × 10⁻⁶. 4 CFU / mL increased significantly to 9.84 × 10⁻⁶. 6 The CFU / mL level increased by two orders of magnitude; while the total bacterial count in the experimental group remained at a low level, slowly increasing from the detection limit (<10 CFU / mL) to 8.68 × 10⁻⁶. 3 CFU / mL. Notably, the total microbial count in the experimental group after 45 days of storage remained lower than the baseline value on day 1 of the traditional group, fully demonstrating the superior storage stability of the experimental group's product during distribution. Although microbial proliferation was inhibited at 4℃, traditional milk tofu still showed a two-order-of-magnitude increase (final value 8.56 × 10⁻⁶ CFU / mL). 4 The experimental group showed a significantly slower microbial proliferation rate (CFU / mL), which is closely related to the antibacterial active substances formed within it. Fresh milk tofu is rich in nutrients, making it an easy substrate for rapid microbial growth and leading to rancidity. The special fermentation agent developed in this study can construct an antibacterial barrier in the milk tofu system through specific metabolic pathways, effectively inhibiting the microbial growth kinetics during storage, keeping the total bacterial count within the safe threshold, thereby simultaneously improving food safety and shelf-life extension.

[0238] The embodiments described above are merely preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Various modifications and improvements made by those skilled in the art to the technical solutions of the present invention without departing from the spirit of the present invention should fall within the protection scope defined by the claims of the present invention.

Claims

1. A fermentation agent for preparing shelf-stable flavored milk tofu, characterized in that, The fermentation agent includes *Lactobacillus pentosacchari* (… Lactiplantibacillus pentosus YN-04 and Lactobacillus plantarum ( Lactiplantibacillus plantarum YN-05; The Lactobacillus pentosaccharide YN-04 was deposited at the China Center for Type Culture Collection (CCTCC) on March 13, 2025, at Wuhan University, Wuhan, China, with accession number CCTCC NO: M2025469. The *Lactobacillus plantarum* YN-05 was deposited at the China Center for Type Culture Collection (CCTCC) on March 13, 2025, at Wuhan University, Wuhan, China, with accession number CCTCC NO: M2025470. The mass ratio of *Lactobacillus pentosolicus* YN-04 to *Lactobacillus plantarum* YN-05 in the fermentation agent is 1:

1.

2. The fermentation agent according to claim 1, characterized in that, The effective viable count of *Lactobacillus pentosacchari* YN-04 in the fermentation agent is ≥10. 9 CFU / mL, the effective viable count of *Lactobacillus plantarum* YN-05 in the fermentation agent is ≥10. 8 CFU / mL.

3. The application of the fermentation agent as described in claim 1 or 2 in the preparation of flavored milk tofu.

4. The application of the fermentation agent as described in claim 1 or 2 in the preparation of shelf-stable milk tofu.

5. A method for preparing milk tofu using the fermentation agent according to claim 1 or 2, characterized in that, Includes the following steps: The raw milk is sterilized, and the fermentation agent is inoculated into the sterilized raw milk. Fermentation is carried out until the pH reaches 4.

6. The whey is removed by heating, the milk is kneaded into a ball, cut, and compressed into shape to obtain milk tofu.

6. The method for preparing milk tofu according to claim 5, characterized in that, The sterilization of raw milk is carried out by placing raw milk in an 80°C water bath, stirring for 15 minutes, and then cooling.

7. The method for preparing milk tofu according to claim 5, characterized in that, The inoculation amount of the fermentation agent is calculated as 3% of the volume fraction of the sterilized raw milk; the fermentation temperature is 25°C.

8. The method for preparing milk tofu according to claim 5, characterized in that, The heating and whey removal process involves a water bath at 65°C.

Images