Optimization of lactic acid bacterial starter culture to improve the quality and flavor characteristics of traditional Hurood

Hurood is a traditional fermented milk product prepared by traditional Mongolian techniques of fermenting raw milk, partial degreasing, heating, whey drainage, emulsification of curd, and molding. Currently, Hurood available in the market is generally prepared by small-scale enterprises at home or in open air. Therefore, lack of standardization of bacterial starter culture leads to variation in the flavor and sensory properties of Hurood from batch to batch. In this study, we aimed to assess the best starter culture combination obtained from 37 lactic acid bacterial strains isolated from traditional Hurood. The solidification state and sensory quality were used as indexes for determining the fermentation efficiency of the bacterial starter culture combinations. The yield and texture characteristics were used to determine the optimal ratio of bacterial strains in a combination and the processing conditions for traditional Hurood production. The most optimal bacterial culture combination was observed to be NF 9–3:NF 10–4:CH 3–1 in 5:4:1 ratio and in 3% amount. The most optimal whey temperature and heating–stirring temperature were observed to be 55°C to 60°C and 85°C to 90°C, respectively. Hurood prepared with the optimal combination of bacterial strains exhibited significantly enhanced sensory quality, flavor, and contents of AA and fatty acids. Therefore, the use of optimal starter culture of lactic acid bacteria could produce Hurood with significantly superior sensory qualities, making the product more acceptable to consumers.


INTRODUCTION
Traditional Hurood is a fermented dairy food product traditionally eaten by nomads in the regions of Inner Mongolia.It is an important source of proteins and contains nutrients that support the growth of human bones and muscles (Machado-Fragua et al., 2020).In addition, the unique, strong milky flavor of traditional Hurood attracts consumers and can be eaten directly, turned into curd, or added to milk tea (Xiao et al., 2016).Hurood is favored by the majority of Mongolian families and is a characteristic ethnic Mongolian dairy product.Traditional Hurood is produced by fermentation of raw milk by microorganisms present intrinsically or extrinsically in the environment; therefore, the quality and flavor of Hurood has certain regional characteristics (Zhao and Li, 2009b).The unique combination of microorganisms in the production of traditional Hurood provides it with a distinct color, through which the product quality can be determined (Na, 2010).
In the process of fermentation, microorganisms increase the nutritional value of dairy products by changing the contents of organic nutrients (Chen, 2013).In industrial production, raw milk needs to be pasteurized to eliminate pathogenic bacteria and some harmful microorganisms, and in this process, some beneficial microorganisms are inevitably inactivated.Supplementation with microbial cultures is a promising strategy to obtain traditional fermented dairy products that meet nutritional and functional requirements.Beneficial microflora is added to raw milk, and consistency in terms of flavor is maintained for fermented dairy products (Zhao et al., 2017).
Auxiliary starter refers to the auxiliary fermentation culture that is exogenously added to the milk when producing fermented dairy products to improve product flavor and speed up fermentation (Tang and Xu, 2017).As the main microbial culture in the production of traditional Hurood, starter culture plays a crucial role in the fermentation.When the starter culture is inoculated into raw milk, the microbial cells grow and multiply under appropriate conditions.Their metabolites have important effects on the flavor, texture, and quality of fermented dairy products.They improve the shelf life, nutritional value, and digestive characteristics of the products.The addition of probiotic culture promotes the growth of beneficial microorganisms and inhibits the propagation of harmful bacteria.
Generally, microbial strains that are beneficial for the production of traditional fermented dairy products are screened using indicators such as fermentation characteristics.For example, Lactobacillus paracasei produces aldehydes, acids, and alcohols and is a useful bacterium for fermentation (Van Hoorde et al., 2010;Stefanovic et al., 2017).Lactic acid bacteria are the most common microorganisms in the production of traditional Hurood, with Lactobacillus, Streptococcus, and Lactococcus being the core bacteria used for the fermentation (Zhao and Li, 2010).Su and Yun (2010) reported that Lactobacillus helveticus could be used as a probiotic, auxiliary starter culture to prepare low-fat cheese, and it imparts beneficial properties to the product.Zhou et al. (2019) reported that the supplementation of auxiliary starter culture greatly shortened the fermentation time, and quality characteristics of the product were improved.Moreover, the content of alcohol significantly increased.
When a single strain is used as an auxiliary starter culture, only a specific type of effect is observed, and the product characteristics will not be significantly affected (Gobbetti et al., 2015).However, when different strains are mixed and used as the starter culture for Hurood, their interaction can promote the formation of various flavor substances and diversification of taste (Xu, 2020).When Lactococcus and Candida are cocultured, they interact to produce diacetyl compounds, thus improving the flavor of fermented products (Lin et al., 2018).Through biochemical analysis and detection of volatile components and free AA, Stefanovic et al. (2018) reported that the addition of 3 different strains of Lacticaseibacillus paracei to skim milk could promote the production of various flavor compounds in cheddar cheese to increase the volatility of the characteristics of the cheese.Li et al. (2019) used Streptococcus thermophilus, Lactobacillus acidophilus, and Lactobacillus germanicus as auxiliary starter culture to prepare Hurood.
Currently, most of the Hurood available in the market is produced by small-scale enterprises, and different varieties of Hurood have not yet been developed.Consumers' perception about Hurood is still relatively limited, because its unique taste may not be acceptable for most people.This is one of the reasons for the poor growth of Hurood sales.Second, as Hurood is mostly produced by small-scale enterprises generally in the home and open-air environment, the constituent microorganisms and their proportions are not standardized.Therefore, the flavor and texture of Hurood is not uniform from batch to batch.In this study, we aimed to identify and optimize the combination of lactic acid bacterial starter culture and fermentation parameters for making traditional Hurood with enhanced sensory quality and flavor, which were assessed using singlefactor and orthogonal tests.This study highlighted the importance of selecting superior lactic acid bacterial starter culture to ferment traditional Hurood and provided a reference for the processing and development of traditional fermented dairy products.

MATERIALS AND METHODS
No human or animal subjects were used, so this analysis did not require approval by an Institutional Animal Care and Use Committee or Institutional Review Board.

Experimental Materials
Thirty-seven experimental lactic acid bacterial strains, isolated from Hurood samples from the Xilingol League area, Inner Mongolia, were provided by the laboratory of the Ethnic Characteristics Research and Development Team of Inner Mongolia Agricultural University, Hohhot, China.Fresh cow milk was supplied by Yide Dairy at Inner Mongolia Agricultural University, Hohhot.Control Hurood samples (commercially available Hurood) were provided by Changhong Dairy Factory, Guxilin Goole League, Zhenglan Banner, Inner Mongolia.

Primary Screening of Strains for Fermentation.
In total, 37 lyophilized and purified lactic acid bacterial strains were inoculated in de Man, Rogosa, and Sharpe (MRS) liquid medium (Guangdong Huankai Microbial Technology Co. Ltd., Guangdong, China) under sterile conditions.The cultures were incubated at 25°C (DHP-9052 Electric Thermostatic Incubator; Shanghai Yiheng Technology Co. Ltd., Shanghai, China) for 24 h and further subcultured 2 to 3 times after incubation at 25°C for 24 h.The bacterial suspension of each strain was inoculated in sterilized skim milk and cultured at 25°C until titratable acidity (°T) of approximately 65°T was obtained.In this way, the parent cultures were obtained and were stored at 4°C.The parent culture of each strain was added to pasteurized whole milk (3% [vol/vol]) for fermentation at 25°C.Solidification status, acid production, aroma properties, and uniform organizational structure were used as the indicators for primary screening.Accordingly, 10 strains were screened and subjected to secondary screening.
Secondary Screening of Strains.After primary screening, the parent culture of screened strains was further inoculated into pasteurized whole milk (3% [vol/vol]) for fermentation at 25°C to prepare traditional Hurood.The strains producing good solidification status, uniform and delicate consistency, and strong aroma were screened.The time required for curd formation, titratable acidity, pH, color, solidification status, and whey precipitation were observed and noted.The time of curd formation, titrated acidity, and sensory and texture characteristics (TMS-ProTA.XT Express; StableMicroSystem Co.) of the product were used as the parameters to screen strains producing smooth curd, strong milky flavor, and high sensory scores.
Growth Characteristics of the Screened Strains.In total, 3 strains were selected in secondary screening.Each of them was inoculated in MRS liquid medium (Guangdong Huankai Microbial Technology Co. Ltd.), incubated at 25°C (DHP-9052 Electric Thermostatic Incubator; Shanghai Yiheng Technology Co. Ltd.), and sampled every 4 h.The samples were placed in a quartz cuvette, and the optical density (OD) was determined 3 times at 600 nm, based on the growth characteristics test of the traditional Hurood starter strain to measure the concentration of bacterial culture medium, so as to estimate the growth of bacteria.
The parent starter culture of strains selected after secondary screening was inoculated into sterilized (SX-500 Steam Sterilization Pot; Tomy Co. Ltd., Japan) whole milk (3% [vol/vol]).The fermentation was conducted at 25°C (DHP-9052 Electric Thermostatic Incubator; Shanghai Yiheng Technology Co. Ltd.), and samples were collected at 0, 4, 8, 12, 16, 20, 24, 28, and 32 h after the inoculation.The viable counts in the samples were measured according to the plate counting method (Milesi et al., 2010).The samples were diluted in various gradients under aseptic conditions.Suitable dilution gradient was inoculated on MRS agar plates and cultured at 37°C (DHP-9052 Electric Thermostatic Incubator; Shanghai Yiheng Technology Co. Ltd.) for 48 h.Plates with the number of colonies between 30 and 300 were selected for total colony count.The titratable acidity and pH were measured 3 times, and the average value was calculated (Liang et al., 2020).

Optimization of Composite Starter Culture of Lactic Acid Bacteria and Fermentation Process.
The parent cultures of 2 strains were mixed in a ratio of 1:1 and inoculated into sterilized whole milk (3% [vol/vol]).Fermentation was conducted in a constant temperature incubator at 25°C.The milk solidification time, acidity, and pH were recorded; color and solidi-fication status were observed, and the combinations of strains with good synergistic effect were screened.
According to the time of curd formation, acidity, and flavor characteristics, various ratios of the parent cultures were designed to prepare starter culture for traditional Hurood.According to the ratio scheme, each single-strain parent culture was added to sterilized whole milk (3% [vol/vol]), and fermentation was conducted at 25°C (DHP-9052 Electric Thermostatic Incubator; Shanghai Yiheng Technology Co. Ltd.).The curd formation time, acidity, and pH were recorded; color and organization state were observed; and the traditional Hurood was prepared.The texture and fermentation characteristics of the products were compared with determine the optimal starter culture combination, and the composite starter culture was prepared.
Once the optimal combination of starter culture was determined, the optimal amount of composite starter culture, whey temperature, and heating-stirring temperature were determined.The single-factor test was conducted to compare the changes in terms of curd formation time, color, and texture under various amounts of composite starter culture, whey temperatures, and heating-stirring temperatures, and their effects on the yield and texture characteristics of traditional Hurood were determined to obtain the most optimal parameter.With the amount of composite starter culture, whey temperature, and heating-stirring temperature as the indexes, the best process parameters of traditional Hurood were determined using an orthogonal test.
(1) Determination of optimal starter culture amount: The selected combination of composite starter culture was added to sterilized whole milk at various amounts (2%, 3%, or 4% [vol/vol]).The whey extraction temperature was set to 65°C, and the heating-stirring temperature was set to 85°C to determine the optimal amount of composite starter culture for traditional Hurood production.
(2) The effect of whey temperature on the quality of traditional Hurood: Using the optimal composite starter culture amount determined earlier and heating-stirring temperature of 85°C, Hurood samples were prepared with the whey extraction temperatures of 55, 60, or 65°C, and the optimal whey extraction temperature was determined.
(3) The effect of heating-stirring temperature on the quality of traditional Hurood: Using the optimal amount of composite starter culture and optimal whey temperature determined earlier, Hurood samples were prepared with the heating-stirring temperatures of 75, 80, and 85°C, and the optimal heating-stirring temperature was determined.
(4) Orthogonal experimental design for traditional Hurood production technology: Based on singlefactor experiment, with sensory evaluation and production rate of control traditional Hurood as the indexes, L 9 (3 3 ) orthogonal experiment was designed.The amount of starter culture, whey expulsion temperature, and heating-stirring temperature were considered as the variables of orthogonal experiment.The orthogonal experiment factors and level design are shown in Supplemental Table S3 (https: / / doi .org/ 10 .6084/m9 .figshare.24312070.v1) to determine the best process parameters for making traditional Hurood.
Determination of the Quality of Hurood.Using commercially available Hurood (Inner Mongolia Xilin Gol League Zhenglan Banner Changhong Dairy Factory) as the control, the influence of lactic acid bacterial starter culture on the characteristics of traditional Hurood was assessed.The quality characteristics (moisture, protein, fat, ash, AA, and fatty acid contents and texture characteristics) were compared with those of the control Hurood sample.
(1) Moisture content determination: Moisture content was determined by referring to GB 5009.2019) with some modifications.In brief, Hurood samples were cut into 2 cm × 2 cm × 2 cm pieces and subjected to texture characteristic determination as described below.The values were determined 3 times, and the average value was calculated.The measurement parameters were TPA mode, probe P/100, pre-measurement velocity 1 mm/s, speed under measurement 1 mm/s, return speed 5 mm/s, form variable 50%, and triggering force 2 g.Each index was evaluated as follows: hardness, the peak of the first rise in the curve; springiness, the ratio of the second projection distance of the curve to the first; cohesiveness, the ratio of the second positive peak area to the first positive peak area; adhesiveness, the first negative peak area of the curve; chewiness, the product of hardness, springiness, and cohesiveness; and recovery, getting back to the original shape even after pressing.
Determination of the Flavor Characteristics of Hurood.The sensory characteristics (odor, taste, and flavor compounds) were compared with those of the control Hurood sample.
(1) Determination of odor: The odor characteristics were determined using the method described by Wang (2019) with some modifications.In brief, Hurood samples were ground in a mortar; 5 g of Hurood powder was loaded into a headspace bottle, and a PEN3 electronic nose (Airsense Co. Ltd.) was used for odor determination at room temperature.The process involved automatic cleaning for 100 s and sample determination for 90 s.Three measurements were collected for each sample, and the average value of data between 73 and 75 s was selected for analysis.(2) Determination of taste: First, Hurood was pretreated as per the method described by Qu (2018) with some modifications.In brief, Hurood samples were ground in a mortar and pestle; 5 g of evenly crushed Hurood samples were mixed in 100 mL of deionized water and centrifuged at 5,589.5 × g and 20°C for 20 min.The supernatant was filtered using a filter paper, and the clear liquid was used for the analysis.The acidic, bitter, astringent, salty, fresh, and sweet sensors and reference electrodes of the SA-402B electronic tongue device (Insent Co. Ltd.) were activated in advance, and the activation time was 24 ± 2 h.The clear liquid was poured into the sample cup for the taste determination.
(3) Determination of flavor compounds: Hurood samples were pretreated as follows.In total, 3 g of each Hurood sample was taken into a 20-mL headspace sample bottle, which was immediately clamped with an aluminum cap with PCE and set aside.The test was conducted in the hydrazine capture mode, and the samples were incubated at 60°C for 30 min.The injection volume was 1 μL, and the samples were analyzed at the injection port for 5 min before analysis using GC-MS.The GC conditions were as follows: the column was DB-wax (30 m × 0.25 mm, 0.25 μm); the carrier gas was helium (99.999%); the flow rate was 1 mL/min; the injection method was splitless injection; the column temperature was held at 40°C for 5 min, followed by an increase to 220°C at 5°C/min and to 250°C at 20°C/min, and held for 2.5 min; and injection temperature was 260°C.The MS conditions were as follows: electron impact ionization ion source; electron energy, 70 eV; ion source temperature, 230°C; interface temperature, 260°C; quadrupole temperature, 150°C; scan mode, full scan; and mass range, 20-400 amu.

Data Processing
Data were analyzed and summarized using Excel software (Microsoft Corp.) for statistics.SPSS version 19.0 software (IBM Corp.) and Past software 4.09 were used to determine significance and data representation.Data were visualized using Origin 2021 (OriginLab Corp.).
Secondary Screening of the Strains.The morphology of the 10 strains was determined.All exhibited milky white and convex colonies.The colonies of all strains were smooth except those of CH 9-1, CH 1-1, CH 2-1, and NF 10-4.Further, the acid production and solidification state of these strains were evaluated (Table 2).The curd formation time for the 10 strains exceeded 32 h, with NF 15-1 exhibiting the longest curd formation time of 62 h.Very short curd formation time is not conducive to the formation of fermented milk flavor.Among the 10 strains, for NF 9-3, the whey was separated and milk flavor was rich.The curd structure of NF 14-3 and CH 4-1 was hard and that of CH 2-1 was soft.Fat in milk is dispersed in whey in the form of pellets with a diameter of 0.2 to 15 μm.Similar to fat pellets, casein exists in the form of colloidal particles, and the structural stability and interaction between particles strongly affect the quality of traditional Hurood (Shima and Tanimoto, 2016).As seen from Figure 1A and 1B, different starter strains exhibited significantly different texture characteristics of Hurood (P < 0.05).Among the 10 strains, the hardness of Hurood produced by NF 12-1, NF 14-3, and CH 3-1 was the highest and that of Hurood produced by CH 1-1, CH 2-1, and CH 9-1 was the lowest.If the hardness of Hurood is too high, the taste is bad.If the hardness is too low, Hurood is too soft and the taste is not very good.Hurood produced by NF 9-3 exhibited the highest viscosity and that by NF 15-1 exhibited the lowest viscosity.The chewiness of Hurood produced by CH 3-1 and NF 10-4 was the highest.No difference was observed among all strains in terms of springiness and recovery (getting back to the original shape even after pressing).
After the secondary screening and texture determination of the 10 lactic acid bacterial strains, 3 strains were finally screened: NF 9-3 with short fermentation and curd formation time, NF 10-4 with good acid production ability, and CH 3-1 with good flavor.The 16S r RNA analysis revealed that NF 9-3, NF 10-4, and CH 3-1 were Pediococcus pentosaceus, Lactobacillus delbruechii ssp.bulgaricus, and Lactococcus lactis, respectively.
Growth Curve and Changes in the Acidity by the Strains.Figure 1 shows the OD 600 of NF9-3, NF10-4, and CH3-1 strains in MRS liquid medium, number of viable bacterial cells in skim milk, and acidity and pH of the fermented skim milk at 25°C.
It can be seen from Figure 1C and 1D that in the same medium, each strain readjusted its metabolic activities to adapt to the new environment.With the extension of culture time, the growth rate of the strains was accelerated; the metabolic activity gradually increased, and the cells entered the logarithmic growth stage.The OD 600 values of NF 9-3, NF 10-4, and CH 3-1 cultures basically stabilized at 12 h, and the number of viable bacteria was 9.48, 7.93, and 8.36 log cfu/mL, respectively.After 12 h, OD 600 values of NF 9-3 culture began to rise slowly.Each strain began to accumulate secondary metabolites, and the number of viable bacteria decreased slowly after 24 h.Both NF 10-4 and CH 3-1 gradually entered a stable period.
As shown in Figure 1E and 1F, the acidity gradually increased and pH gradually decreased during the fermentation of skim milk.Among the 3 strains, the titratable acidity of NF 10-4 culture increased faster than that of the other 2 strain cultures during the whole fermentation period, and finally, the titratable acidity of NF 10-4 culture could reach 83°T.The titratable acidity of NF 9-3 and CH 3-1 cultures increased slowly in the first 8 h but rapidly in the late fermentation period, reaching 83°T and 79°T, respectively.In the early stage of fermentation, the pH varied slightly, ranging from 6.24 to 6.60, which might be due to the strains' self-regulation of metabolic processes to adapt to the new environment.The pH of NF 9-3 and CH 3-1 cultures decreased steadily for 4 to 20 h and reached 4.62 and 4.71, respectively, at the end of fermentation.The pH of NF 10-4 culture reached 5.24 at 12 h and decreased slowly, reaching 4.50 at the end of fermentation.The acid production capacity of the 3 strains followed the order NF 10-4 > NF 9-3 > CH 3-1.

Optimization of the Combination of the Strains.
To determine the optimal combination of the strains, the dominant strains were combined in a 1:1 ratio and used to produce Hurood.As shown in Supplemental Figure S2 (https: / / doi .org/ 10 .6084/m9 .figshare.24312070.v1), the 1:1 combination of NF 9-3, NF 10-4, and CH 3-1 could form Hurood, and no antagonism was observed among the 3 strains.Moreover, the curd was smooth and delicate and had the flavor of traditional Hurood.Therefore, further experiments were conducted: NF 9-3 with faster curdling, NF 10-4 with better acid production, and CH 3-1 with good fermentation flavor were fermented in different proportions.The parameters of different proportions are shown in Supplemental Table S4 (https://doi.org/10.6084/m9.figshare.24312070.v1).
Optimization of the Fermentation Process. Figure 2 shows the effects of various starter culture amounts (2%, 3%, and 4%), whey temperatures (55°C, 60°C, and 65°C), and heating-stirring temperatures (80°C, 85°C, and 90°C) on the sensory score and yield of traditional Hurood.The sensory score was the highest under 3% composite starter culture amoun (Figure 2 (a)).Increasing or decreasing the starter culture amount may lead to under-or overfermentation of traditional Hurood, resulting in poor taste.When the starter culture amount reached 3%, the yield was the maximum (9.7%).However, further increase in the starter culture amount would lead to the loss of protein and other solids during the whey discharge process, resulting in curd insufficiency and decrease of the yield.The most important role of starter culture in the preparation of traditional Hurood is to produce acid.The difference in starter culture amount directly affects the curd formation process, thus changing the texture characteristics of traditional Hurood.The hardness of traditional Hurood gradually increased with the increase in the amount of starter culture and changed from 4,572.65 to 4,838.55 g (Figure 2 (b)).This may be because with the increase in the amount of starter culture, the speed of acid production increased rapidly, which resulted in the accumulation of a large amount of whey, increase in the hardness, and formation of a film on the surface (Zhao and Li, 2009a).The springiness, adhesiveness, and cohesiveness of the product were the best when the starter culture amount was 3%.Therefore, 3% starter culture amount was observed to be the most suitable for the preparation of traditional Hurood.
Whey expulsion temperature will affect the shrinkage of clots and exudation rate of whey liquid, which is one of the important factors controlling the moisture content of traditional Hurood.Very high or very low whey expulsion temperature would hinder the continuous propagation of lactic acid bacteria to a certain extent and would prevent the continuous decline in acidity, leading to the poor taste formation of traditional Hurood (Zhao et al., 2010).With the increase of whey expulsion temperature, the sensory score and yield of traditional Hurood first increased and then decreased (Figure 2C).When the whey expulsion temperature was 60°C, the sensory score reached the maximum value, and the yield was the highest.Gel properties are very important in traditional Hurood production to trap fat droplets in the protein network and promote the contraction of clots after cutting; therefore, whey expulsion temperature is significantly correlated with fat loss in whey (Panthi et al., 2019).The hardness of traditional Hurood increased with the increase of whey expulsion temperature and reached the highest value of 5,231.06g at 65°C (Figure 2D).The higher the whey expulsion temperature is, the more the whey is discharged and the less the moisture content is in the final traditional Hurood product.The best whey expulsion temperature to prepare traditional Hurood was observed to be 60°C.
The effect of heating-stirring temperature on the yield of traditional Hurood was not obvious (Figure 2E).Under all heating-stirring temperatures, the yield was above 8%.At 90°C, the sensory score and yield of traditional Hurood was the lowest, which may be because the heating-stirring temperature is closely related to hardness.When the temperature is too high or too low, the curd of traditional Hurood cannot form in a good state.After heating and stirring at an optimal temperature, the texture becomes rough, which indirectly leads to the decrease in the taste quality (Aldalur et al., 2019).Heating-stirring temperature clearly affected the hardness of Hurood (Figure 2F).With the increase of temperature, hardness value gradually decreased; it ranged from 4,071.899 to 4,772.65 g.The chewiness reached the lowest value when heating-stirring temperature was 85°C, when the springiness of Hurood was the highest.Springiness is an important factor to assess the quality of traditional Hurood (Vacca et al., 2019a).Therefore, the optimal heating-stirring temperature was determined to be 85°C with sensory evaluation and yield as the main indexes, along with texture characteristics.
Orthogonal Analysis of Traditional Hurood Production.The orthogonal test results of traditional Hurood are given in Table 3.The orthogonal test results with sensory score as the index indicated that the optimal parameter combination was A2B2C2 (i.e., the starter culture amount of 3%, whey extraction temperature of 60°C, and heating-stirring temperature of 85°C).The orthogonal test results with the yield as the index indicated that the best parameter combination was A2B1C3 (i.e., the starter culture amount of 3%, whey extraction temperature of 55°C, and heating-stirring temperature of 90°C).
Physical, Chemical, Flavor, and Quality Characteristics of Traditional Hurood Fermented by Lactic Acid Bacterial Starter Culture.The protein components of Hurood play an important role in milk solidification and shrinkage, and the change in the relative content of specific protein components has a huge impact on the quality of traditional Hurood (Amalfitano et al., 2019;Vacca et al., 2019b).The pH value and protein content of Hurood prepared with starter culture were 4.64 ± 0.21 and 31.62% ± 0.53%, respectively (Figure 3A).The protein content was higher by 0.86% than that of the control group (30.76% ± 0.16%; P < 0.05), and the ash content (1.68% ± 0.06%; P < 0.05) was higher by 0.15% than that of the control group (1.53% ± 0.03%; P < 0.05).Excessive heating-stirring temperature and long duration would lead to severe moisture loss and increased DM content in the Hurood prepared with starter culture.

Contents of AA and Fatty Acids in the Experimental and Control Hurood Samples.
The AA content in traditional Hurood can be regarded as an indicator of the biochemical activity of lactic acid bacteria.Proteins generate polypeptides and AA under the action of hydrolytic enzymes; thus, the AA content reflects the protein content of traditional Hurood.In cow milk with higher casein content, the curd formation requires shorter time, and curd is firmer and easier to shrink; thus, it is conducive to the more uniform discharge of whey (Franciosi et al., 2011) and does not lead to the loss of protein along with the expulsion of whey.As shown in Figure 3B, 16 types of AA were detected in both traditional Hurood, and the total amount of AA in the experimental samples was 30.61 g/100 g sample, which was higher by 7.18 g/100 g sample than that in the control samples.Some lactic acid bacteria or their metabolic pathways have high nutritional requirements; thus, the strains will consume some AA.The content of glutamic acid decreased, which was 0.96 g/100 g sample and 4.91 g/100 g sample, respectively, in experimental and control samples.The content of glutamic acid decreased by 5.11 times, and the content of other AA was higher in the experimental samples than that in the control samples.Some AA contribute to the improvement in the sensory properties of tradi-tional Hurood (Ge, 2008).For example, methionine is an essential AA, which is the main precursor of the flavor of traditional Hurood.The increase in methionine content will promote the formation of flavor substances in the fermented product.The methionine content in the experimental samples was higher than that in the control samples.The contents of tyrosine and histidine were approximately 1.78 times higher than those in the control samples, which were the 2 AA with the highest increase in the experimental samples followed by threonine, serine, phenylalanine, and proline with approximately 1.76 times increase compared with the control samples.The content of AA in the other 8 experimental samples was approximately 1.50 times more than that in the control samples; the content of aspartate was 1.40 times more than that in the control samples, reaching 2.56 g/100 g sample.Some studies reported that the content of aspartate increased in Hurood prepared with added Lactobacillus, indicating that aspartate may be the source of flavor-enhancing substances (Zheng et al., 2012).In general, the AA content of traditional Hurood was significantly improved by adding different proportions of lactic acid bacteria.
As seen from Figure 3C, 23 fatty acids were detected in the experimental and control samples, with no difference in the types of fatty acids.In terms of content, the relative percentage of these 23 fatty acids was slightly lower in the experimental samples than in the control samples, and the content of SFA was higher by 2.6% in the experimental samples than in the control samples.The relative content of lauric acid, myristic acid, stea- K1 is the number of levels of each factor in the experiment.K2 is the square of the number of levels for each factor in the experiment.K3 is the cube of the number of levels for each factor in the experiment.R-value refers to the fourth power of the number of levels for each factor in the experiment.ric acid, oleic acid, and linolenic acid was significantly lower in the experimental samples than in the control samples.The relative content of oleic acid in the experimental samples was 19.59%, which was less by nearly 3.05% than that in the control samples.Palmitoleic acid content was almost the same in both the types of samples.The relative content of quillic acid, myridamic acid, pentacarbonate, palmitic acid, and linoleic acid was significantly higher in the experimental samples than in the control samples.It was reported (Xu et al., 2019) that linoleic acid in Hurood could produce substances with unique aroma under the action of enzymes.The relative content of palmitic acid in the experimental samples was 34.81%, which was higher by nearly 2.38% than that in the control samples.In addition, the content of short-chain fatty acids (number of carbon atoms less than 6 in the carbon chain) in the experimental samples was the least, accounting for only 2.31% of the total AA, whereas the content of mediumchain fatty acids (number of carbon atoms 6-12 in the carbon chain) accounted for 10.04% of the total AA.Long-chain fatty acids (more than 12 carbon atoms in the carbon chain) accounted for 87.63% of the total AA.It was speculated that long-chain fatty acids might be one of the components responsible for the typical aroma of Hurood.
Odor and Taste of Hurood Fermented by Lactic Acid Bacterial Starter Culture.The flavor of dairy products produced by different fermentation processes will be different.In this study, electronic nose and electronic tongue were used to analyze the odor and taste of the experimental and control Hurood samples (Figure 4).
As seen from Figure 4A, in terms of odor, the experimental samples were significantly different from the control samples.Among them, the response values of W1C, W3C, and W5C sensors significantly increased in the experimental samples.W1C is sensitive to aromatic compounds, mainly detecting C 7 H 8 functional group; W3C is sensitive to aromatic component, and W5C is sensitive to aliphatic aromatic compounds.The response values of these 3 sensors increased in the experimental samples.It indicated that the aroma of the experimental Hurood samples was better than that of the control samples.The response values of W5S, W6S, W1S, and W2S sensors did not change significantly.The response values of W1W, W2W, and W3S sensors decreased significantly in the experimental samples compared with the control samples.These 3 sensors are mainly sensitive to organic sulfides and alkanes, indicating that the addition of lactic acid bacteria during Hurood production had a significant effect on the reduction of organic sulfides, quality, and aromatic substance content (Wei et al., 2015).
As seen from Figure 4B, in terms of taste, the experimental samples were different from the control samples.Particularly, significant differences existed in terms of sour taste, richness, aftertaste-B, aftertaste-A, sweet taste, and salty taste between the 2 samples.However, no significant difference existed in terms of bitter, astringency, and umami.The acidity, richness, aftertaste-B, aftertaste-A, and sweetness of the experimental samples were lower than those of the control samples.This could be attributed to the difference in the microbial communities during the production process.As the input of environmental microorganisms was larger in the control samples than in the experimental samples, the fermentation flavor and response value were too strong in the control samples during the detection.Saltiness was most likely because of high mineral content (Yang et al., 2021).
The principal component analysis of flavor quality of the experimental and control Hurood samples based on electronic nose and electronic tongue is shown in Figure 4.The first principal component (PC1) consisted of 6 indicators detected using sensors such as W1W, W6S, W1C, W3S, W2S, and W1S, accounting for 83.18% of the weight of all variables (Figure 4C).The second principal component (PC2) consisted of indicators detected using 4 sensors, including W5S, W5C, W3C, and W2W, accounting for 16.75% of the weight of all variables.In PCA factor score chart, the experimental and control Hurood samples exhibited a relatively clear clustering trend and obvious separation (Figure 4D); therefore, the odor quality of the experimental and control Hurood samples was different.Figure 4E) shows that the contribution rate of PC1 was 90.01%, which was composed of 5 indexes including sour taste, aftertaste-A, aftertaste-B, umami taste, and saltiness.The contribution rate of PC2 was 6.63%, which was composed of 4 indexes: sweetness, astringency, bitterness, and abundance.Figure 4F factor score chart indicated that a significant separation existed between the experimental and control samples, indicating that the taste quality of the 2 Hurood samples was different.
Analysis of Flavor Characteristics of Hurood.Fermented milk has a unique flavor.Table 4 shows the corresponding flavor thresholds.The OAV (the ratio of flavor substance concentration to its threshold) ≥1 indicates that the substance contributes to aroma components in fermented products; however, OAV ≥10 indicates that the substance contributes to important aroma components in fermented products.As seen from Table 4, 16 substances with OAV ≥10 were observed in the experimental and control samples; however, the OAV of the experimental samples was greater than that of the control samples, indicating that the flavor of the experimental samples was better, Analysis of Contents of Flavor Substances in Hurood.The GC-MS was performed to determine the volatile flavor substances in the experimental and control Hurood samples.In total, 52 and 42 flavor substances were detected in the experimental and control samples, respectively.The flavor substances were well separated, and the species were abundant in the experimental group (Figure 5A and 5B).
The flavor of the experimental samples was stronger than that of the control samples (Figure 5A), indicating that the lactic acid bacterial starter culture improved the flavor of the product.In general, acids dominate the flavor, followed by alcohols, lipids, alkanes, aldehydes, and others.
Acids are important flavor substances of traditional Hurood, and free fatty acids are representative flavor components of acids.As shown in Figure 5A and 5B, the total acid content of the experimental samples was 18,923 mg/100 g sample; the major acids were benzoic acid, butyric acid, isobutyric acid, valerate acid, heptanic acid, caprillic acid, caproic acid, and capric acid.The total acid content of the control samples was 3,006.709mg/100 g sample.The high level of acids in the experimental samples may be attributed to the enhanced conversion of decomposable enzymes by the auxiliary starter, which leads to accelerated production of acidic flavor substances (Weng et al., 2019).Different acids bring different flavors to Hurood.Studies have reported that the content of caprylic acid and capric acid can affect the milky flavor of products, and the higher the content, the stronger the milky flavor (Liu et al., 2008).A total of 10 alcohols were detected in the experimental samples, with amount of 3,137.76mg/100 g sample, whereas 6 alcohols were detected in the control samples, with amount of 973.2 mg/100 g sample.The number of alcohols in the experimental samples was higher, possibly because the contents of fatty acids and AA in the experimental samples were higher than those in the control samples.Alcohols are flavor substances, with sweet or banana aroma, and are obtained by oxidation or degradation of fatty acids (Niu et al., 2010).
A total of 11 esters were detected in the experimental (2,001.476mg/100 g) and control (404.9774mg/100 g) samples.Lipids can be decomposed into aromatic substances, which are highly volatile under high temperature conditions and have a unique milky flavor (Ma et al., 2013).
A total of 7 aldehydes were detected in the experimental (1,118.717mg/100 g) and control (293.07 mg/100 g) samples.Among them, acetaldehyde was present in abundance.According to Zhao et al. (2008), acetaldehyde greatly affects flavor of products.Aldehydes are mainly produced by the degradation of AA and are important flavor substances in fermented milk (Madruga et al., 2009).

CONCLUSIONS
In this study, we optimized the lactic acid bacterial starter culture and fermentation process by determining the best starter culture ratio and amount, whey temperature, and heating-stirring temperature to produce traditional Hurood with enhanced sensory characteristics and quality.The most optimal bacterial culture combination was observed to be NF 9-3:NF 10-4:CH 3-1 in 5:4:1 ratio and in 3% amount.The most optimal whey temperature and heating-stirring temperature were observed to be 55-60°C and 85-90°C, respectively.The results indicated that the sensory qualities, flavor, and amino acid and fatty acid contents of Hurood produced with optimal lactic acid bacteria starter culture and fermentation characteristics were significantly improved.The texture of Hurood produced with the lactic acid bacterial starter was denser than the commercial Hurood.Therefore, the use of lactic acid bacterial starter culture produced Hurood that was significantly superior to the commercial Hurood in terms of sensory characteristics, making the product more acceptable to consumers.Inner Mongolia Agricultural University (Grant No. BR220137; Hohhot, China), Xilin Gol League Science and Technology Plan (Grant No. 202117;Hohhot), and Inner Mongolia Central Government Guide Local Science and Technology Development Fund Project (Grant No. 2022YFXZ0032;Hohhot).I thank my teacher, corresponding author Quan Shuang, for his guidance in the process of writing the paper.The authors have not stated any conflicts of interest.

Figure 1 .Figure 2 .
Figure 1.Strain quality characteristics: texture characteristics of secondarily screened strains (A and B); optical density (OD) at 600 nm in de Man, Rogosa, and Sharpe medium (C), viable cells in skim milk medium (D), and acidity (E) and pH (F) of fermented milk.Error bars indicate SD of 3 independent experiments.a-d: Within a row, means with different letters are significantly different (P < 0.05).

Figure 4 .
Figure 4. Odor and taste of Hurood fermented by lactic acid bacterial starter culture: radar map of response values of electronic nose (A) and tongue (B) in the experimental and control Hurood sample.Factor load diagram and factor score diagram of Hurood odor quality (C, D); factor load diagram and factor score diagram of Hurood taste quality (E, F).PC = principal component.
Yang et al.:  LACTIC ACID BACTERIAL STARTER CULTURE TO IMPROVE HUROOD and the addition of auxiliary starter culture improved the flavor of Hurood.

Figure 5 .
Figure 5. Contents of flavor substances in Hurood: classification and content of flavor chemicals in the experimental and control samples (A); heat map of flavor substance content of Hurood (B).Cls = classification; CH = control Hurood; NF = experimental Hurood.
Yang et al.: LACTIC ACID BACTERIAL STARTER CULTURE TO IMPROVE HUROOD Yang et al.: LACTIC ACID BACTERIAL STARTER CULTURE TO IMPROVE HUROOD 3-2016 "Determination of moisture in food"(Food  Partner Network, 2021)and using the direct drying method.(2)Determination of protein (KDY-9830 Kjeldahl Nitrogen Analyzer; Shanghai Sheng Sheng Automated Analytical Instrument Co. Ltd.) content: Overall, 0.5 g of Hurood samples were weighed and ground into fine particles.Further they were used to determine protein content by referring to the first method of the national standard GB 5009.5-2016(Kaynitrogendetermination method; Food Partner Network, 2021)."Determination of ash in food" (Food Partner Network, 2021).(5)Determination of AA (Biochrom 30+ Automatic Amino Acid Analyzer; BioChrom Co. Ltd.,

Table 2 .
Changes in acidity and solidification state during fermentation Yang et al.: LACTIC ACID BACTERIAL STARTER CULTURE TO IMPROVE HUROOD

Table 3 .
Yang et al.: LACTIC ACID BACTERIAL STARTER CULTURE TO IMPROVE HUROOD Orthogonal test results of traditional Hurood process optimization

Table 4 .
Yang et al.: LACTIC ACID BACTERIAL STARTER CULTURE TO IMPROVE HUROOD Flavor composition and odor activity value (OAV) of the experimental and control samples