Functional properties and flavor characteristics of milk from cows supplemented with jujube powder

Jujube has various functional properties and is a promising source of bioactive compounds and flavors. This study investigated the functional properties and flavor characteristics of milk from cows supplemented with jujube powder ( JP). Here, milk volatile profiles and taste properties were analyzed by using an electronic nose, and headspace solid-phase microextraction gas chromatography-mass spectrometry (HS-SPME-GC-MS). Compared with the control group, the total antioxidant capacity, 2,2 ′ -azino-bis-3-ethylbenzothi- azoline-6-sulphonic free radical scavenging activity, lactoferrin, and IgG levels increased significantly in the JP group. Volatile flavor analysis indicated that ketone levels increased, while acid abundance decreased, and toluene and dimethyl sulfone significantly increased in the JP group. Taste profile analyses demonstrated that jujube supplementation altered the taste of the milk. In summary, dietary jujube powder supplementation affects the volatile flavor composition and aroma of milk, as well as the bioactive component and antioxidant properties. These findings enhance our understanding of milk production using direct dietary supplementation to produce sustainable dairy products.


INTRODUCTION
Milk is rich in protein, calcium, antioxidants, immune globulins and other functional components (Wang et al., 2018), thus it is generally beneficial to human health.In recent years, consumer demand for health-enhancing foods has increased substantially owing to the desire for improved quality of life.Bioactive substances and flavors are 2 key indicators for evaluating milk qual-ity (Superti, 2020).Functional studies have revealed that the bioactive components in milk, such as bovine lactoferrin, immunoglobulins, and antioxidants, have physiological benefits beyond basic nutrition (Ulfman et al., 2018, Khan et al., 2019, Kell et al., 2020).Milk flavor is one of the most important characteristics associated with consumer acceptability.Traditionally, milk is processed into various dairy products and beverages by adjusting processing or adding ingredients, such as sugar and fruit pulp, to satisfy various groups of consumers (Feng et al., 2019).However, these increase production costs and health concerns because of the added sugar, which has been linked to cardiovascular disease and lower nutrient value (Yang et al., 2014).
Dairy products containing "natural" ingredients have the highest level of acceptance among consumers (Bimbo et al., 2017).As a novel green approach, dietary supplementation of dairy cows with natural foods has great potential in the milk industry.This process not only has positive benefits on animal health (Shen et al., 2009), but can also improve milk flavor and increase the content of bioactive compounds in milk through nutrient metabolism (Clarke et al., 2019, Debong andLoos, 2020).Diets can impact milk composition and flavor in dairy cows (Bendall andChemistry, 2001, Clarke et al., 2019).The source of flavor substances in raw milk is affected by the direct transfer of aroma compounds from feed to milk, and the formation of aroma compounds during feed digestion (Carpino et al., 2004).
Several studies have focused on the importance of the Chinese jujube (Ziziphus jujuba Mill.), a plant in the Rhamnaceae family that has various biological effects, including antioxidant, immunomodulatory, and antitumor effects (Ji et al., 2017).The multiple beneficial effects of jujube can be attributed to its nutritional and chemical components, including polysaccharides (Wu et al., 2021), vitamins, flavonoids (Pawlowska et al., 2009), and nucleosides (Gao et al., 2013).However, few studies have investigated the effects of dietary supplementation with jujube powder (JP) on the regulation of milk flavor and bioactive compounds.Thus, we hypothesized that supplementing dairy cows with jujube could affect Functional properties and flavor characteristics of milk from cows supplemented with jujube powder Chen Zhang, 1 Jie Mei, 1 Yinxiang Wang, 2 Bo Yu, 2 and Hongyun Liu1* their milk flavor and functional components.In this study, the functional properties and flavor differences of milk under different conditions (diets with and without jujube powder) were investigated.Our study provides a theoretical foundation for further applications in natural milk production.

Animals, diets and experimental design
The experiment was conducted at a modern largescale dairy farm in Heze City, Shandong Province.The experimental procedures were approved by the Animal Care Committee at the Zhejiang University (Hangzhou, China) and were under the University's guidelines for animal research.The jujube from the same batch was dried, and then the jujube fruits were crushed by the crusher 40 mesh sieve to a particle size ≤ 0.5 mm.Jujube powder was mixed with other ingredients to total mixed ration diet (TMR).Forty healthy multiparous Chinese Holstein dairy cows with a similar day in milk (159.41 ± 34.5 d, mean ± SD), and milk yield (38.26 ± 3.18 kg) were randomly allocated into 2 blocks with 20 cows in each block.Blocks were randomly assigned into 2 treatments, basal diets without (control, CT) or with jujube powder (JP) at 400 g/d.The experiment consisted of a 2-week acclimatization period and an 8-week data collection period.

Milk composition analysis
Milk composition (protein, fat, lactose, total solids, and somatic cell count [SCC]) was analyzed using an infrared analysis system with a 4-channel spectrophotometer (Milk Foss-4000, Foss Electric A/S, Hillerod, Denmark).

E-nose analysis
The electronic nose, an effective way to analyze aroma, can provide an overview of flavor (Tiggemann et al., 2017).Milk aroma profiles were characterized using a Fox 4000 electronic nose (e-nose; Alpha M.O.S.; Toulouse, France).The sensor array was composed of 18 metal oxide sensors of 3 types with various preferred foci.Each milk sample (2 mL) was placed into a 10-mL rubber stoppered vial and incubated for 30 min at 40°C.Processed pure air (Hangzhou Jingong Special Gas Co., Ltd.) was used as a carrier gas to clean the sensor array and to zero the signal response, after which the sample was measured using an injection volume of 2 mL.For each sample, 4 biological replicates and 2 technical replicates were measured, and the mean value was used for further analysis.The electronic nose system was preconditioned and calibrated according to the procedure recommended by the manufacturer.The airflow rate was set at 150 mL/min.The data acquisition time was 120 s for each sample and the clean phase was 240 s.The sensors were cleaned and standardized before and after detection to eliminate drift.

Volatile compounds analysis by HS-SPME-GC-MS
The aromatic compounds in JP and milk were analyzed using headspace solid-phase microextraction gas chromatography-mass spectrometry (HS-SPME-GC-MS).Milk sample (5 mL) and 2-Octanol (internal standard, 10 μL, 10 mg/L) were transferred into a 20 mL headspace vial.After incubation at 60°C for 15 min, the volatiles were then sampled with a 100-mm sorbtive extraction (SE) probe fused silica fiber coated with DVB/CAR/PDMS (Sigma-Aldrich) for 30 min.After extraction, the fiber was immediately inserted into the injection port of the GC-MS for the desorption step at 250°C for 4 min.GC-MS analyses were performed using an Agilent 7890 gas chromatograph system coupled with a 5977B mass spectrometer.The chromatographic separation was achieved on a DB-WAX capillary column (30 m × 250 μm × 0.25 μm).Helium was used as a carrier gas at a flow rate of 1 mL/min through the column.The oven temperature was maintained at 40°C for 4 min, raised to 245°C at a rate of 5°C/min, and maintained for 5 min.The injector, transfer line, ion source, and quadrupole detector temperatures were set to 250, 250, 230, and 150°C, respectively.The MS electron impact mode was recorded at 70-eV ionization energy.Detection was performed in full scan mode over Zhang et al.: Functional properties and flavor… a mass m/z range of 20-400, and the solvent delay was to as 0 min.Qualitative identification of the compounds were performed using the Chroma TOF 4.3X software (LECO Corporation) by comparing their mass spectra and retention times with the standard spectra provided by the National Institute of Standards and Technology (NIST 11) database (Agilent Technologies Inc., Gaithersburg, MD, USA).The retention indices of each volatile compound were calculated using the retention time.Compounds were quantified by integrating the selected specific ions to obtain relative peak areas.

Statistical analysis
For bioactive ingredients, e-nose responses, and milk volatile compounds data were analyzed using oneway ANOVA (ANOVA), followed by Duncan's test to identify significant differences.Principal component analysis (PCA) was conducted performed using the "ggbiplot" package in R. Spearman's rank correlation analysis was performed to evaluate the potential link between the alterations in serum parameters, e-nose responses, and volatile compounds in milk.Significance was declared at P ≤ 0.05, and 0.05 < P ≤ 0.10 was considered statistically significant.

Volatile compounds in jujube powder
The aromatic compounds found in jujube powder are showed in Table S1.The type and content of volatile compounds directly affect the flavor of milk, which is a very important property affecting the acceptability and preference of humans.In total, 122 volatile compounds were identified, including 19 acids, 15 ketones, 15 alcohols, 13 lactones, 7 hydrocarbons, 18 aldehydes, 22 esters, 7 pyridines, 3 furans, and 3 sulfur-containing compounds.Many of these compounds have been identified as volatiles in previous jujube studies (Pan et al., 2022).Our results showed that formic, butanoic, and pentanoic acid were the main acids in JP.Acids were the dominant compounds in this study, which is consistent with other studies on dried jujube (Song et al., 2020), followed by ketones, alcohols and lactones.These volatile compounds of jujube have many positive effects on aroma formation, and the use of jujube to produce natural foods is a better alternative for overcoming health risks (Feng et al., 2019, Wang et al., 2019, Yang et al., 2022).

Alterations in milk composition and active substances
The concentrations of the main components and bioactive substances in milk are shown in Table 1.Compared with the CT group, there was an increasing trend in milk protein content (P = 0.06) in the JP group.No differences were found in fat content (P = 0.55), lactose content (P = 0.49), or total solid content (P = 0.26) between the 2 dietary groups.In addition, the SCC was not changed by dietary supplementation with JP.However, T-AOC (P < 0.01), ABTS (P = 0.02), lactoferrin (P < 0.01), and IgG (P < 0.01) levels increased significantly in the JP group.The increase in T-AOC and ABTS levels showed that jujube promoted the production of antioxidants in milk.The content of lactoferrin and IgG in milk was significantly higher in the JP group than in the CT group.IgG is derived from blood and is transferred to mammary secretory cells (Gapper et al., 2007).Therefore, our results suggest that jujube powder contributes to immunoglobulin synthesis.Bovine lactoferrin has been extensively studied owing to its antimicrobial and immunomodulatory properties (Superti, 2020).The mechanism underlying this process may be lactoferrin directly binding to intestinal tissues and bacteria because lactoferrin receptors have been found in intestinal tissues and in some bacteria (Legrand et al., 2006).However, methods for improving lactoferrin content in milk have not been thoroughly investigated.Our findings indicate that feeding jujube to dairy cows increases lactoferrin content in milk.

Influence on volatile organic compounds in milk
The preservation of volatile compounds in milk from dairy cows directly affects human acceptance.The relative concentrations of aroma compounds in milk are listed in Table 2.For milk flavor, the principal components (PCs) PC1 and PC2 represented 56.96% and 13.48% of the total variance, respectively, which explains 70.44% of the total variance.A significant difference was observed between most of these compounds (Figure 1A, 1B, and 1C).A total of 56 volatile organic compounds related to flavor were detected in the milk samples.The volatile compounds were classified into 10 groups based on their chemical properties: 10 ketones, 11 acids, 9 aldehydes, 4 alcohols, 4 esters, 7 lactones, 6 hydrocarbons, 3 furans, one pyridine, and one sulfurcontaining compound.The most abundant compounds detected in milk in this study were acids, ketones, and hydrocarbons, in agreement with Natrella et al. ( 2020), in which milk was characterized by a higher ketone and acid content.The main acid and ketone compounds in the milk differed (P < 0.05) between the CT and Zhang et al.: Functional properties and flavor… JP groups (Figure 1D).The results show that volatile organic compounds (such as butanoic acid, 2-butanone, and dimethylsulfone) exhibited significant differences (P < 0.05) after dietary supplementation with JP.
Acids in milk.In this study, butanoic and heptanoic acid levels were decreased (P < 0.05) in the JP group.Decanoic acid accounted for a small proportion of aroma compounds but was significantly increased in JP group.Previous studies have shown that decanoic acid could alleviate inflammatory cytokine production and oxidative stress (Mett and Müller, 2021).The alteration in volatile acids may be attributed lipolysis of fat (Perez-Palacios et al., 2010).Therefore, dietary supplementation with JP may alter the metabolic pathways of milk, which may affect the flavor characteristics of milk.
Ketones and alcohols in milk.Ketones and alcohols are primarily formed through biochemical processes involving triglycerides lysis and saturated fatty acid oxidation.The relative concentration of ketone was higher in the JP group than that in the CT group.In this study, the main ketones were 2-heptanone and 2-butanone, with higher levels of 2-butanone detected in the JP group compared with the CT group.Notably, 2-butanone contributes to floral and fruity aromas (Zhu et al., 2021).Ketones, specifically acetone and butanone, are naturally present in dairy products, and derive from bovine metabolism or feed (Amador-Espejo et al., 2017, Lemos et al., 2022).There was an increasing trend of 1-octen-3-ol (P = 0.053) in the JP group than the CT group, which was associated with a grassy odor.Alcohols may also originate from the reduction of corresponding aldehydes or amino acid metabolism (Yue et al., 2015) and their contribution to the VOC profile of milk has a high odor threshold (Amador-Espejo et al., 2017).
Aldehydes in milk.Nine aldehyde compounds were identified in milk, with pentanal and heptanal being the predominant compounds.In the present study, octanal content (P = 0.012) was higher in the JP group, whereas decanal (P < 0.01) and benzeneacetaldehyde (P = 0.021) contents were lower than in the CT group.Octanal is responsible for the characteristics of citrus oranges (Miyazawa et al., 2010).Aldehydes have lower threshold values than alcohols and can impart a significant aroma, either pleasant or rancid, to foodstuffs (Ma et al., 2013).Thus, although aldehydes are present in low amounts, they can override other flavoring substances.Generally, aldehydes are produced by the degradation of amino acids or lipid oxidation in milk (Clarke et al., 2020).Therefore, the alteration in milk aldehyde content may be associated with a JP-supplemented diet.
Hydrocarbons in milk.In this study, toluene and p-xylene were the predominant hydrocarbons, with toluene levels being significantly higher (P < 0.05) in JP than CT milk.Toluene is associated with sweetness and can modulate immunity.In a previous study, exposure of infant mice to 50 ppm toluene enhanced total plasma immunoglobulin levels (Yamamoto et al., 2009).The p-cresol has the antioxidant activity (Pan et al., 2017) and this might account for the increased antioxidant capacity of milk.Toluene and p-cresol are both degradation products of β-carotene (Faulkner et al., 2018) and potential biomarkers for milk derived from natural pastures, which have higher values because of their perceived healthiness and environmental acceptability.This may be due to the fact that the JP diet altered hydrocarbon production, which can subsequently be transferred to the milk.
Sulfur-containing compounds in milk.Higher amounts of dimethyl-sulfone were found in the JP milk than in the CT milk (P < 0.01).Dimethyl sulfone has anti-inflammatory and reactive oxygen species scavenging actions (Santos et al., 2003), which may be due to the presence of methionine and cysteine (Villeneuve et al., 2013).Our result agrees with Coppa et al. (2011), in which a higher relative concentration of dimethylsulfone was found in pasture-derived milk than in the total mixed-ratio diet.Studies show that sulfur-containing compounds, such as dimethyl disulfide, hydrogen sulfide, dimethyl trisulfide, and methanethiol, were identified in milk.Our study detected only dimethyl sulfone, likely because of limitations in the sensitivity of the mass spectrometric detector.This prevented the analysis of the sulfur-containing compounds present at lower relative concentrations.The alteration in dimeth-  yl sulfone levels in milk may result from the JP diet, which could potentially impact methionine metabolism.Overall, our results show that volatile organic compounds in milk are affected by dietary supplementation with JP.The potential underlying mechanisms could be tested on a large scale in a wide variety of production environments.Thus, the relationship between the aroma characteristics of milk and aroma substances should be further studied.Further research is necessary to understand the dynamics of variations in milk volatile organic compounds concentrations during diet changes.

Differential aroma release profiles of milk by E-nose
The PCA results of the response values of the e-nose are presented in Figure 2A.The principal components PC1 and PC2 represented 69.48% and 17.75% of the total variance, respectively, with the cumulative contribution of the first 2 PCs accounting for 87.23%, indicating that they were sufficient to explain the total variance (Wold et al., 1987).The variance contribution of PC1 was greater than that of PC2, indicating that the greater the distance along the PC1 axis, the greater the sample variation.PCA analysis indicated that the e-nose responses to different milk samples exhibited significant distinctions, implying variations in milk aroma.This finding aligns with the results obtained from SPME-GC-MS analysis.A radar plot was used to visualize whether the pattern differed between milk samples.As shown in Figure 2B, there were significant differences and similarities between the fingerprints.The variation trends of signals from different samples were similar, with the T30/1, P10/1, P10/2, P40/1, T70/2, PA/2, P30/1, T40/2, T40/1, and TA/2 sensors having stronger responses to volatile compounds in milk, indicating higher abundances of organic, hydrocarbon, and alcohol compounds.Heatmap cluster analysis was also used to assess the contribution of the e-nose sensors to the discrimination of similarities and dissimilarities among the 8 milk samples, as shown in Figure 2C.The milk samples were categorized into 2 clusters: CT-1, CT-2, CT-3, and CT-4 comprised one cluster, while JP-1, JP-2, JP-3, and JP-4 constituted the other.A unique combination of volatile flavor compounds was observed between the 2 groups.This result agrees with the separation of aroma characteristics of these 2 types of milk according to e-nose signals.

Correlation between serum parameters, e-nose responses, and milk volatile compounds
Correlation analysis was performed to analyze the relationship between milk aroma compounds, serum parameters, and e-nose measurements.The results reveal that the sensors exhibited different responses to different volatiles compounds (Figure 3B).Acids, ketones, and hydrocarbons are among the key aroma compounds in all milk samples; this is shown by their correlation with the majority of the e-nose sensor results.The e-nose has a series of sensors designed to detect volatile flavors (Coloretti et al., 2014).The signals of the TA/2, T40/1, PA/2, T70/2, P40/1, P10/2, P10/1, and T30/1 sensors were positively correlated with acids but negatively correlated with ketones and sulfur-containing compounds (P < 0.05).Lactones were positively correlated with P30/2 (P < 0.05), whereas aldehydes were not significantly correlated with any other sensors.Additionally, alcohols were negatively correlated with LY2/gCTL, and esters were positively correlated with T30/1 and LY2/LG.Hydrocarbons were negatively correlated with P30/2 and LY2/AA.Furan was positively correlated with LY2/AA and pyridine was positively correlated with LY2/gCT and LY2/LG.Briefly, the volatile components and aroma detected by the sensors can make combinative contributions to the differentiation of milk flavors in this study.The aroma compounds in milk can also be formed through a series of metabolic processes in vivo, in which different metabolites in the serum affect the flavor of milk in the mammary glands.Therefore, a correlation analysis between the volatile compounds in milk and serum parameters was performed (Figure 3A).Acids, esters, and lactones in milk were positively correlated with CHOL, AST, and MDA levels in the serum (P < 0.05).Lactones were negatively correlated with IgG and T-AOC  levels in this study, indicating that the increase of the antioxidant and immune functions in dairy cows may lead to a reduction in lactone content.
Although the untargeted analysis of volatile compounds was performed in the current experiment, the targeted volatile compounds for the absolute quantification need to be assessed in future research.Moreover, sensory evaluation is still required to further compare the sensory perception of raw milk between CT and JP cows.

CONCLUSIONS
Dietary jujube powder supplementation was effective in (1) increasing the content of lactoferrin and IgG in milk; (2) increasing total antioxidant capacity and free radical scavenging activity; and (3) affecting the flavor characteristic of the milk by increasing certain volatile compounds, such as 2-butanone and Octanal.These findings enhance our understanding of organic milk production using direct dietary supplementation of dairy cow diets with natural foods to achieve sustainable dairy production.

Figure 1 .
Figure 1.Alterations in volatile compounds of milk after dietary supplementation with jujube powder.(A) PCA and (B) volcano plot of volatile compounds between CT and JP groups; (C) cluster heatmap analysis upon the classification of the samples and volatile compounds; (D) relative content of ketones and acids between CT and JP groups.

Figure 3 .
Figure 3. Correlations between (A) volatile compounds in milk and differential serum parameters (P < 0.05), and (B) E-nose sensors.Each row in the graph represents a volatile compound, each column represents a serum parameter and sensor, and each lattice represents a Spearman correlation coefficient.Red represents a positive correlation, while blue represents a negative correlation.

Table 2 .
Relative concentrations of aroma compounds in milk of CT (n = 4) and JP (n = 4) groups by SPME-GC-MS Zhang et al.: Functional properties and flavor…

Table 2 (
Continued).Relative concentrations of aroma compounds in milk of CT (n = 4) and JP (n = 4) groups by SPME-GC-MS