Effect of a mixture of Lactobacillus sakei and Staphylococcus xylosus on the fermentation quality of chicken breast meat

Abstract

BACKGROUND

Meat products made through microbial fermentation show significant market prospects due to their unique flavor, texture, and long shelf life. However, research on inoculation of fermentation into chicken breast meat is still poorly understood. This study investigates the effects of Lactobacillus sakei and Staphylococcus xylosus, applied individually and in varying combinations, on the fermentation of chicken breast. The research examines changes in physicochemical properties, free amino acid content, and flavor profiles of the fermented chicken breast samples.

RESULTS

Based on single‐factor and orthogonal experiments, the optimal processing conditions were a starter ratio of L. sakei M2/S. xylosus Y4 = 2:1, a total inoculum quantity of 1 × 10⁷ CFU g⁻¹, fermentation at 30 °C for 24 h, and a final pH of 5.12. Compared with the uninoculated control, starter‐inoculated samples showed significantly lower moisture content, water activity, hardness and L* and b* values (P < 0.05), whereas springiness, chewiness, a* value, volatile flavor diversity and sensory scores significantly increased. Among all treatments, the 2:1 mixed‐culture group produced the most desirable overall quality, with enhanced redness, richer volatile profiles and higher levels of taste‐active free amino acids.

CONCLUSION

Mixed fermentation with L. sakei M2 and S. xylosus Y4 effectively improved the texture, color, flavor and sensory acceptability of fermented chicken breast meat. The combination at a 2:1 ratio showed the greatest potential as a starter culture for the development of fermented chicken breast products. © 2026 Society of Chemical Industry.

INTRODUCTION

Chicken breasts are highly nutritious and rich in high‐quality protein, potassium, phosphorus, iron, zinc, and several B vitamins. ¹ , ² In China, chicken meat attracts many consumers owing to its rich nutritional value; however, poor taste is the primary factor affecting chicken consumption. ³ Current approaches to improve chicken meat quality include ultrasonication, electrical stimulation, and food additives. A recent study found that marinating chicken with glutathione significantly increased the amount of flavor substances that impart a barbecue flavor to chicken. ⁴ Ultrasonic treatment in low‐salt chicken paste induced protein conformational changes, which improved chicken emulsion stability and gelation. ⁵ However, the processing of prepared chicken breast involves several steps. Moreover, physical and chemical treatments can adversely affect the quality and flavor of chicken breasts and reduce product shelf life.

Fermented meats are widely appreciated for their distinctive taste, extended shelf life, and numerous nutritional benefits, including probiotic properties. ⁶ Lactic acid bacteria, coagulase‐negative staphylococci, yeasts, and molds are commonly utilized as fermenting agents to enhance the overall quality of meat products. ⁷ Lactobacillus sakei demonstrates competitive ability against other bacteria during fermentation, while Staphylococcus xylosus can enhance the redness of meat and contribute to the formation of fermentation flavors. ⁸ , ⁹ Multi‐strain probiotics have the potential to modulate gut microbiota when consumed, possibly offering health benefits beyond basic nutrition. ¹⁰ , ¹¹ Multi‐strain fermentation is known to produce more complex flavors. A study examined the relationship between community succession and aroma in traditional Chinese fermented fish and identified 81 aroma compounds in fish fermented with mixed cultures, which contributed to improved aroma quality. ¹²

In the preliminary investigation, our research team isolated a strain of L. sakei M2 from bacon and a strain of S. xylosus Y4 from a meat fermenter. The optimal growth conditions for L. sakei M2 and S. xylosus Y4, both individually and in combination, were determined for fermented chicken breast production. These conditions were established through the examination of inoculum quantity, fermentation temperature, and fermentation duration. Subsequently, the study explored the quality changes in chicken breast meat fermented with L. sakei M2 and S. xylosus Y4 under these optimized conditions, laying the groundwork for future research on fermented chicken breast products.

MATERIALS AND METHODS

Materials and chemicals

Lactobacillus sakei M2 and S. xylosus Y4 obtained from the College of Food and Bioengineering at the Hefei University of Technology (Hefei, Anhui, China) were isolated and identified by our research group. ¹³ Fresh white chicken breast and ingredients such as sugar, salt, spices, monosodium glutamate, and cooking wine were procured from a local market (Anhui RT‐Mart, Hefei, Anhui, China). The Amino Acid Standard (AAS18) was purchased from SIGMA Co., Ltd, Beijing, China. The standard of 2‐methyl‐3‐heptanone (99%) was obtained from LGC Biosearch Technologies Co., Ltd, Beijing, China. Sodium ascorbate was acquired from Siyang Biotechnology Co., Ltd (Jining, Shandong, China).

Bacterial activation

For the microbial preparation L. sakei M2, isolated from Sichuan bacon, and S. xylosus Y4, obtained from meat starter cultures, were cultivated in De Man–Rogosa–Sharpe (MRS) broth and nutrient broth medium, respectively. Following two generations of incubation, the bacteria were inoculated into seed medium and incubated at 37 °C for 12 h. Subsequently, 1% (v/v) of the bacterial solution from the seed medium was transferred to the amplification medium, and incubation continued at 37 °C for 12 h to prepare the fermentation starter. ¹⁴ For mixed‐culture treatments, freshly prepared L. sakei M2 and S. xylosus Y4 suspensions were combined at the required volumetric ratio immediately before inoculation. Unless otherwise specified in the experimental design, the mixed starter used in the single‐factor tests was prepared at a 1:1 ratio.

Process recipe

The formulation for 1 kg of chicken breast was as follows: sodium chloride, 10 g kg⁻¹; vanilla, 0.3 g kg⁻¹; monosodium glutamate, 0.3 g kg⁻¹; glucose, 5 g kg⁻¹; cooking wine, 10 g kg⁻¹; sodium nitrite, 0.15 g kg⁻¹; sucrose, 40 g kg⁻¹; and sodium ascorbate, 0.5 g kg⁻¹.

Sample preparation

1. Preparation of fermentation broth: The bacterial broth was centrifuged at 7000 × g for 10 min. Preparation of starter suspension: the cultures were centrifuged, the supernatant was discarded, and the cell pellets were resuspended in an equal volume of sterile 0.85% saline to prepare the starter suspensions required for the different treatments. The bacterial suspension was then prepared into the required fermentation agent according to the production specifications. ¹³

2. Pretreatment of raw meat (at 4 °C): The chicken breasts were removed from their packaging and thoroughly rinsed under running water to ensure all surfaces were properly cleaned.

3. Marination: The prepared chicken breasts were fully immersed in the marinade and refrigerated at 4 °C for 6 h. Fermentation: Seven points were identified on the surface of the chicken breasts, and the bacterial solution was injected into these points using a disposable needle. Subsequently, the chicken breasts were placed in a constant‐temperature‐and‐humidity incubator and fermented at 30 °C for 24 h.

4. Steaming: Following marination and fermentation, the chicken breasts were steamed in a steamer basket for 10 min.

5. Baking: The fermented chicken breast was baked in a drying oven at 50 °C for 2 h, with the meat being rotated at 30 min intervals to ensure uniform heating and prevent adhesion to the surface.

6. Packaging of finished products: Following the cooling of baked and cooked products, vacuum packaging was employed to preserve the color and luster of the products, inhibit fat oxidation, and facilitate storage at 0–4 °C.

One‐way analysis of variance

To determine the starter culture strain ratio fermentation was conducted at 30 °C for 24 h with a total starter inoculum of 1 × 10⁷ CFU g⁻¹. Six groups of single factor tests were designed to determine the optimal ratio. Control group (CK): samples not inoculated with fermenter; L. sakei M2 single group (3:0): samples inoculated with L. sakei M2 only; L. sakei M2/S. xylosus Y4 combined group (2:1); L. sakei M2/S. xylosus Y4 combined group (1:1); L. sakei M2/S. xylosus Y4 combined group (1:2); S. xylosus Y4 single group (0:3): samples inoculated with only S. xylosus Y4.

To determine the total starter inoculum quantity fermentation was carried out at 30 °C for 24 h with L. sakei M2/S. xylosus Y4 = 1:1. The total inoculum quantity levels were 1 × 10⁵, 1 × 10⁶, 1 × 10⁷, 1 × 10⁸ and 1 × 10⁹ CFU g⁻¹.

To determine the fermentation temperature with the mixed‐strain ratio fixed at L. sakei M2/S. xylosus Y4 = 1:1 and fermentation time set at 24 h, the total starter was inoculated at 1 × 10⁷ CFU g⁻¹ and fermentation was carried out at 20, 25, 30, 35 or 40 °C.

To determine the fermentation time with the mixed‐strain ratio fixed at L. sakei M2/S. xylosus Y4 = 1:1, fermentation temperature set at 30 °C and total starter inoculum at 1 × 10⁷ CFU g⁻¹, chicken breast samples were fermented for 20, 25, 30, 35 or 40 h.

The optimal combination for the chicken breast fermentation process

Drawing from the outcomes of the one‐way test, specific factors and levels were identified for subsequent orthogonal tests.

The pH analysis

Following the method described in Janssens et al., ¹⁵ with a minor modification, 5.0 g of samples was thoroughly mixed in 50 mL of 0.1 mol L⁻¹ potassium chloride solution for 2 min. The pH levels were then measured using a pH meter (PHB‐5; Shanghai INESA Instrument Co., Ltd, Shanghai, China). Three independently prepared samples were analyzed for each treatment (n = 3), with each sample measured in triplicate (k = 3).

Sensory evaluation

Sensory panel

Pre‐made chicken breasts were evaluated in a controlled sensory room by a panel of ten trained food industry professionals. Each panel member assessed pre‐made chicken breasts subjected to various treatments. ¹⁶ To facilitate sensory recovery, a 10 min break was scheduled between each round of assessments. This study was reviewed and approved by the Institutional Review Board (IRB) of Anhui Science and Technology University, and informed consent was obtained from each member of the sensory panel prior to their participation in the study.

Sensory scoring criteria

The participants were instructed to evaluate the color, aroma, and chewiness of the pre‐made chicken breasts using a point scale (Table 1).

Table 1.

Chicken breast sensory scoring criteria

Index	Marking scheme	Value of a score
Organizational status (20 points)	Dense tissue, elastic, suitable hardness	16–20
	Tissue slightly loose, slightly sticky, moderate hardness	10–15
	Tissue loose and hardened, tissue adhesion, no elasticity	<10
Appearance and color (20 points)	Brownish red, brownish yellow, uniform color	16–20
	Basically brownish red, brownish yellow, uneven color	10–15
	Blackish or whitish, non‐uniform color	<10
Odor (30 points)	Fermented aroma, no odor	21–30
	Faint fermented aroma, no odor	10–20
	Harsh and other strange aroma	<10
Taste (30 points)	Easy to chew, moderately firm, salty or bland	21–30
	Easy to chew, slightly hard, slightly salty, or slightly bland	10–20
	Dry and hard, with obvious sour or rotten aroma	<10

Moisture activity and water activity analysis

The moisture content of chicken breast was determined by hot‐air drying; the calculation formula is given in Eqn (1). Water activity was directly measured at 25 °C using a water activity meter (RH‐1000; Guangzhou Runhu Instruments Co., Ltd, Guangzhou, China). Three independently prepared samples were analyzed for each treatment (n = 3), and each sample was measured in triplicate (k = 3). ¹⁷ Samples were dried at 105 °C for 20 h, and then weights were recorded every 2 h until the weight difference was < 0.2 mg.

$$w\%=\frac{m_0-m_1}{m_0}\times 100$$ (1)

where m ₀ represents the initial weight of the chicken sample before drying (in grams), and m ₁ represents the weight of the chicken sample after drying (in grams).

Texture profile analysis

Texture profile analysis (TPA) was performed using a TA‐XT Plus Texture Analyzer (Stable Micro Systems, Godalming, UK) equipped with a P/36R probe. Cubes (1 cm × 1 cm × 1 cm) were cut from each treatment sample. Six cubes from each treatment were analyzed (n = 6). The test conditions were as follows: pre‐test speed 2 mm s⁻¹, compression ratio 50%, and trigger force 5 g. ¹⁸

Color analysis

The method outlined in Chaosap et al. ¹⁹ was adapted to assess the color characteristics of the samples. A colorimeter (HP–C210; Shanghai Spectrum Optoelectronics Technology Co., Ltd, Shanghai, China) was employed for calibration using a black‐and‐white plate. The standard light source was then positioned perpendicular to the sample surface for measurement. Color parameters were expressed as L*, a*, b*, and C* values. Three independently prepared samples were analyzed for each treatment (n = 3), and three readings were taken at different positions on each sample surface (k = 3).

Electronic‐nose (E‐nose) analysis

Samples (5 g each) were placed into 50 mL vials and equilibrated at 25 °C for 30 min. Subsequently, a PEN3 electronic nose (E‐nose) system (Airsense Analytics, Schwerin, Germany), equipped with ten metal–oxide semiconductor sensors (Table 2), was used for testing. The self‐cleaning time and detection time were set to 80 s each. The sensor signal values recorded between 69 and 71 s were selected for data analysis. Three independently prepared samples were analyzed for each treatment (n = 3).

Table 2.

A PEN3 electronic nose (E‐nose) sensor sensitive substances

Sensor number	Sensor code	Sensitive materials
1	W1C	Aromatic compound
2	W5S	Nitrogen oxide
3	W3C	Ammonia, aromatic molecule
4	W6S	Hydride
5	W5C	Olefins, aromatics, polar molecules
6	W1S	Alkane
7	W1W	Sulfur compound
8	W2S	Alcohols, some aromatic compounds
9	W2W	Aromatic compounds, organic compounds of sulfur
10	W3S	Alkanes and aliphatics

Solid‐phase microextraction gas chromatography–mass spectrometry (SPME–GC–MS) analysis

The gas chromatography–mass spectrometry (GC–MS) analysis was conducted as described in Liu et al., ²⁰ with minor modifications. A sample of 3.0 g and 1.5 μL of 2‐methyl‐3‐heptanone were combined in 20 mL vials. The mixture was then subjected to extraction (50 °C, 40 min) and desorption (250 °C, 3 min) using a 7890B‐7000C Agilent Technologies instrument (Agilent Technologies, Santa Clara, CA, USA) equipped with a DB‐5MS column (30 mm × 250 mm × 250 mm). Three independently prepared samples were analyzed for each treatment (n = 3).

The experimental conditions for the DB‐5MS column included a flow rate of 1.0 mL min⁻¹, with 99.99% helium as the carrier gas. The column temperature was initially maintained at 40 °C for 3 min and then gradually increased to 250 °C at a rate of 5 °C min⁻¹. Subsequently, the temperature was held at 230 °C for 12 min.

The content of the volatile compound (C _v) was calculated using Eqn (2):

$$\begin{array}{c}{C}_{\mathrm{v}}=\frac{S_{\mathrm{v}}}{S_{\mathrm{a}}}\times C_{\mathrm{a}}\end{array}$$ (2)

where S _v and S _a represent the peak areas of the volatile compound and 2‐methyl‐3‐heptanone, respectively; C _a denotes the concentration of 2‐methyl‐3‐heptanone.

Free amino acid analysis

The free amino acid content was determined according to the method presented in Li et al., ²¹ with slight modifications. Briefly, 2.0 g of each sample was combined with 6 mL of 5% sulfosalicylic acid, and the sample was left at room temperature for 15 min, followed by ultrasonic extraction for 45 min. The resulting mixture was centrifuged at 11 100 × g for 15 min, and 2 mL of the supernatant was collected. Afterward, 1 mL of this supernatant was filtered through a 0.22 μm mixed cellulose ester membrane. The free amino acid content was then analyzed using an amino acid autoanalyzer (Baikang Amino Acid Analyzer; Baikang Co., Ltd, Waterbeach, Cambridgeshire, UK). Three independently prepared samples were analyzed for each treatment (n = 3). The taste activity value (TAV) was calculated to assess taste contribution: TAV > 1 indicates a significant contribution to taste presentation, while TAV < 1 suggests a minimal impact on taste.

$$\begin{array}{c}\mathrm{TAV}=\frac{C}{T}\end{array}$$ (3)

where C represents the free amino acid mass concentration in g kg⁻¹, and T denotes the amino acid taste value in g kg⁻¹.

Statistical analysis

Unless otherwise stated, three independently prepared chicken breast samples from each treatment group were used as biological replicates (n = 3). For instrumental measurements, each biological replicate was analyzed in technical triplicate (k = 3), and the mean value of the technical replicates was used for statistical analysis. For TPA, six independently cut cubes from each treatment were measured, as described later.

The results were presented as the mean ± standard deviation. Basic data processing was performed using SPSS software. Differences were considered significant at P < 0.05 in the analysis of variance. Origin 2021 software was utilized to draw graphs, while Excel 2013 was employed for table creation. Partial least squares‐discriminant analysis (PLS‐DA) was conducted using SIMCA 14.1, and heat maps were generated using TBtools.

RESULTS AND DISCUSSION

Optimization of total inoculum quantity, fermentation time and fermentation temperature

The impact of fermentation duration on pH and sensory scores of fermented chicken breast meat is illustrated in Fig. 1(a), which depicts a gradual decline in pH throughout the fermentation process. The initial 0–12 h period exhibited a rapid pH decrease, indicative of accelerated growth and multiplication of fermenting bacteria, primarily L. sakei M2. This strain consumed carbohydrates in the meat marinade, metabolizing them to produce lactic acid as the predominant organic acid. ²² , ²³ The pH reduction rate decelerated during the 12–30 h period, potentially owing to the progressive depletion of nutrients, resulting in slower metabolic activity of the fermenting strains. Concurrently, sensory scores demonstrated an initial increase followed by a subsequent decrease as fermentation time extended. The initial improvement in sensory scores corresponded to increased metabolic activity of the strains. However, the subsequent decline in scores was due to excessive acid accumulation, rendering the chicken breast meat overly acidic and beyond the acceptable range for sensory appreciation.

Figure 1.

Effects of fermentation conditions on pH, sensory score, moisture content and water activity of chicken breast. (a) Effect of fermentation time on pH and sensory score; the mixed starter ratio was Lactobacillus sakei M2/Staphylococcus xylosus Y4 = 1:1 and the total inoculum quantity was 1 × 10⁷ CFU g⁻¹ at 30 °C. (b) Effect of total inoculum quantity on pH after 24 h fermentation; the mixed starter ratio was 1:1 at 30 °C. (c) Effect of fermentation temperature on pH and sensory score after 24 h fermentation; the mixed starter ratio was 1:1 and the total inoculum quantity was 1 × 10⁷ CFU g⁻¹. (d) Moisture content and water activity of chicken breast under different starter treatments fermented under the optimal conditions. Data are presented as mean ± standard deviation.

Figure 1(b) illustrates the impact of inoculum quantity on the pH of fermented chicken breast meat, revealing significant differences (P < 0.05) in final pH values among the groups. In the 10⁵ group, the pH remained at 5.66 after 24 h owing to insufficient inoculum, failing to reduce to below pH 5.2 within the given timeframe. A rapid pH decrease during fermentation is beneficial for inhibiting harmful microorganism growth, reducing protein water‐binding capacity, and accelerating the drying process. Low pH is crucial for ensuring the safety of fermented meat products. ²⁴ The fermentation groups exhibited significantly lower pH values as the inoculated L. sakei produced acid rapidly under appropriate temperature and humidity conditions. After 24 h of fermentation, the pH values for the 10⁶, 10⁷, 10⁸, and 10⁹ CFU g⁻¹ groups were 5.18, 5.05, 4.91, and 4.70, respectively, all below pH 5.2. This pH reduction is attributable to the rapid acid production by the inoculated L. sakei M2 during the fermentation stage. However, in groups with larger inoculum amounts, nutrients might have been consumed during the pre‐fermentation period, potentially limiting further growth of the fermenting organisms. ²⁵

The fermentation temperature significantly influences the pH and sensory scores of chicken breasts, as illustrated in Fig. 1(c). Lower scores at 20 °C were attributed to poor growth of fermenting strains at this temperature. At 30 °C, the pH was approximately 5.133; however, sensory scores declined when temperatures exceeded this point, possibly owing to reduced protease production by S. xylosus affecting the fermented aroma. ²⁶ The temperature had a great influence on the sensory evaluation, and the sensory score increased with the increase of temperature, and reached the peak at 30 °C, the higher the temperature, the stronger the fermentation effect, and the score increased and then decreased, with a significant difference between the two (P < 0.05). The pH value decreased with the increase of temperature, and the pH value at 40 °C was close to 4.533, which was attributed to the acid‐producing effect of the lactobacilli. The pH values were 5.133, 4.833, and 4.533 at 30, 35, and 40 °C, respectively, all of which were within the safe range (pH 5.2).

Orthogonal experiment analysis

To examine the effects of processing technologies, varying concentrations of inoculum (10⁶, 10⁷, and 10⁸ CFU g⁻¹) were employed to regulate the pH (≤ 5.2). A combination of treatment temperatures (25, 30, and 35 °C) and durations (18, 24, and 30 h) was applied to improve fermentation efficiency and induce quality degradation. Orthogonal experiments were subsequently designed to further optimize these conditions (Table 3).

Table 3.

Factor levels for orthogonal test L9 (3⁴)

Level	Factor
	A: Inoculum quantity (CFU g−1)	B: Fermentation temperature (°C)	C: Fermentation time (h)
1	106	25	24
2	107	30	30
3	108	35	36

Sensory evaluation provides critical information on the sensory qualities of chicken meat, including appearance, flavor (aroma and taste), texture, and the overall acceptability of the product. ²⁷ To determine acceptable process parameters for the fermentation technique of chicken breast, orthogonal tests were conducted based on pre‐experiments. Range analysis results indicated that fermentation temperature was the most crucial factor in the fermentation technique, followed by inoculum quantity and fermentation time (R_B > R_A > R_C). All factors, except fermentation time, demonstrated statistically significant effects on the sensory evaluation of fermented chicken breasts (P < 0.05). The optimal level combination A₂B₂C₁ (fermentation time = 24 h, fermentation temperature = 30 °C, and inoculum quantity = 10⁷ CFU g⁻¹) was identified based on the highest mean response (K _i) value of each optimal level. Three parallel replications were performed, yielding a mean sensory score of 94.0 (Table 4).

Table 4.

Results of orthogonal test L9 (3⁴)

Source of variation	Equation of squares	Optima level	Mean square	F‐Value	Significance
Revised model	323.700	6	53.950	46.732	0.021
Intercept	62917.361	1	62917.361	54500.120	<0.001
A	144.429	2	72.241	62.553	0.016
B	174.562	2	87.281	75.604	0.013
C	4.709	2	2.354	2.039	0.329
Inaccuracies	2.309	2	1.154
Total	63243.370	9
Corrected total	326.009	8

Note: Data are presented as mean ± standard deviation (n = 3). Different lowercase letters indicate significant differences (P < 0.05). R ² = 0.993 (R ² _adj = 0.972).

Impact of various treatments on TPA and moisture content in chicken breast

TPA was employed to simulate the chicken chewing process and determine the textural parameters of the samples, including hardness, elasticity, cohesion, and chewiness. ²⁸ Table 5 illustrates the impact of different treatment conditions on the TPA parameters of chicken. The springiness and chewiness of chicken breast meat in the fermentation group were significantly higher than those in the control group (P < 0.05), suggesting that the decrease in pH caused protein denaturation to form gel tissue, which enhanced the elasticity of chicken breast meat. This finding aligns with the results reported in Purslow ²⁹ and Zhang et al. ³⁰ The meat's textural properties are primarily influenced by the type and structural proteins and the changes in these proteins during meat processing. ³¹ , ³² The hardness was significantly lower than that of the control group (P < 0.05), indicating that the addition of the fermentation agent substantially reduced the hardness of the chicken meat. This effect was attributed to both strains of bacteria producing proteases, which could contribute to the decomposition of myoplasmic and myofibrillar proteins in the chicken meat and disrupt the internal bonding of the meat product. The springiness of the fermentation group was higher than that of the control group, which is consistent with the observed moisture content results.

Table 5.

Effects of different treatments on texture profile analysis (TPA) in chicken breast

Sample	CK	3:0	2:1	1:1	1:2	0:3
Hardness (g)	2785.27 ± 220.57a	1700.25 ± 159.67c	1778.62 ± 192.56c	1863.51 ± 181.62bc	1979.83 ± 229.91b	2029.83 ± 203.29b
Springiness	0.53 ± 0.03d	0.56 ± 0.02c	0.64 ± 0.03a	0.59 ± 0.03b	0.60 ± 0.01b	0.57 ± 0.03c
Chewiness (g)	5179.64 ± 214.2d	6614.37 ± 278.22c	7725.51 ± 261.52a	7323.49 ± 236.49b	6909.41 ± 171.15bc	6073.23 ± 179.05c

Note: Data are presented as mean ± standard deviation (n = 6). Different lowercase letters indicate significant differences (P < 0.05). CK, control group: samples not inoculated with fermenter; 3:0, Lactobacillus sakei M2 single group: samples inoculated with L. sakei M2 only; 2:1, L. sakei M2/Staphylococcus xylosus Y4 combined group; 1:1, L. sakei M2/S. xylosus Y4 combined group; 1:2, L. sakei M2/S. xylosus Y4 combined group; 0:3, S. xylosus Y4 single group: samples inoculated with only S. xylosus Y4.

Low moisture content and water activity effectively inhibit the growth of foodborne pathogenic bacteria. ³³ Figure 1(d) demonstrates that both L. sakei M2 and S. xylosus Y4 possessed strong protein hydrolysis capacity, enabling them to grow and multiply on muscle fiber surfaces. The breakdown of protein molecules between muscle fibers reduced bonding energy and the water‐holding capacity of polar groups on protein molecule surfaces. The moisture content of the control group was significantly higher than that of the fermentation group (P < 0.05). Protein denaturation, resulting from acid production by the fermenter during fermentation, led to decreased water‐holding capacity and subsequently reduced water activity. ¹⁰

Effect of various treatments on the color of pre‐prepared chicken breast

The color of meat products significantly influences consumers' purchasing decisions, as they often rely on visual assessment when shopping. ³⁴ The L* value of the fermented group was significantly lower than that of the control group (P < 0.05) (Table 6), indicating darker meat color, which potentially enhances consumer purchase intent. In fermented chicken breast production, the redness value a* is the most critical color parameter. The a* value of the fermented group was significantly higher than that of the control group (P < 0.05). This can be attributed to two factors: the addition of S. xylosus Y4 to form nitrites and L. sakei M2 reacting with nitrites to produce NO, which combined with myoglobin to form nitroso‐myoglobin and nitroso‐hemoglobin, resulting in a brighter red color. ⁸ The b* value of the fermented group showed a significant decrease (P < 0.05) compared with the control group, suggesting that the fermentation agents reduced the yellowness of fermented chicken breast meat. While the redness value a* was most critical for color in fermented chicken breast, the yellowness value b* and the brightness value L* also contributed to the overall color perception. These findings align with those of Bottegal et al.'s study. ³⁵

Table 6.

Effect of Lactobacillus sakei and Staphylococcus xylosus on the color of fermented chicken breasts

Sample	CK	3:0	2:1	1:1	1:2	0:3
L*	83.11 ± 0.84a	75.16 ± 0.62d	77.50 ± 0.66c	78.95 ± 0.69bc	79.75 ± 0.70b	80.02 ± 0.73b
a*	4.42 ± 0.36e	4.78 ± 0.50d	6.10 ± 0.57a	5.80 ± 0.59b	5.23 ± 0.49c	4.65 ± 0.52d
b*	16.69 ± 0.61a	15.01 ± 0.70b	11.24 ± 0.46d	12.43 ± 0.47c	13.98 ± 0.55bc	14.36 ± 0.68b

Note: Data are presented as mean ± standard deviation (n = 3). Different lowercase letters indicate significant differences (P < 0.05). CK, control group: samples not inoculated with fermenter; 3:0, L. sakei M2 single group: samples inoculated with L. sakei M2 only; 2:1, L. sakei M2/S. xylosus Y4 combined group; 1:1, L. sakei M2/S. xylosus Y4 combined group; 1:2, L. sakei M2/S. xylosus Y4 combined group; 0:3, S. xylosus Y4 single group: samples inoculated with only S. xylosus Y4.

The E‐nose analysis

The significance of various volatile components to the overall aroma profile varies. ³⁶ The analysis of volatile organic compounds (VOCs) in food matrices for quality control largely relies on E‐nose analyses. ³⁷ These devices, which simulate the human olfactory system, are extensively employed for VOC analysis in food. The E‐nose radargram (Fig. 2(a)) revealed notable distinctions between the test and control groups and significant differences among the various inoculum ratios. This suggests substantial variations in the volatile strengths and proportions of gases across different combinations. The six samples elicited strong responses from the W1W, W2S, and W1S sensors (Table 2), indicating that these compounds may exert a greater influence on the samples' volatile profiles. Across all samples, W6S, W3C, and W5C responses were low and not significantly different, suggesting that all samples contained fewer hydrides, ammonia, and aromatic compounds, aligning with Guo et al.'s findings. ³⁸ Furthermore, the 2:1 sample set potentially contained higher concentrations of terpenes, sulfides, alcohols, and alkanes than the other samples.

Principal component analysis for E‐nose findings

The relationship between sample variation and E‐nose data was more effectively elucidated via principal component analysis (PCA), which accurately differentiated chicken breasts treated with various fermenters. Figure 2(b),(c) illustrates the PCA results. The first two principal components (PC1 and PC2) accounted for 80.6% of the total variance in the original data, demonstrating their significant explanatory power. According to the PCA algorithm, samples were categorized into six distinct groups, as depicted in Fig. 2(b). The analysis successfully distinguished chicken breasts subjected to different fermenter treatments. The control sample and the 3:0 sample were positioned in the right quadrant, showing a positive association with W1C. The 1:2, 1:1, and 2:1 samples were located in the upper left quadrant, although not close to each other, indicating distinctive aroma profiles among these samples. These three samples exhibited positive associations with W1W, W5S, W3S, and W2S, which were sensitive to hydrocarbons, ammonia, and aromatic compounds. The 0:3 sample was in the lower left quadrant, demonstrating a positive association with W2W, which is sensitive to aromatic hydrocarbons.

Figure 2.

Electronic nose (E‐nose) dataset of differently treated chicken breasts: (a) radar plot; (b) score plot; (c) loading plot.

SPME–GC–MS analysis of aroma compounds

Volatile characteristics are crucial determinants of food aroma, particularly during chicken breast fermentation, significantly influencing consumer preferences. ³⁹ The analysis detected 72 volatile components in control and fermented chicken breast samples, comprising eight alcohols, 12 aldehydes, seven ketones, seven esters, one furan, seven acids, ten terpenes, 13 alkanes, and three aromatic compounds (Fig. 3(a)). The 2:1 samples exhibited the highest number of VOCs, surpassing that of the control sample, indicating that mixed bacterial fermentation samples possessed more diverse aroma profiles. Furthermore, variations in the quantities of each VOC type across different samples revealed distinct aroma characteristics among them.

Figure 3.

Solid‐phase microextraction gas chromatography–mass spectrometry (SPME–GC–MS) volatile organic compound (VOC) analysis of six samples: (a) number of VOCs in chicken breast before and after fermentation; (b) venn diagram of the number of VOCs in chicken breast with mixed bacteria; (c) heat map of VOCs of chicken breasts with different treatments.

For the control and composite fermenter‐added samples, 28 VOCs were detected, including 3,7‐dimethyl decane, 4‐methyl dodecane, 5‐butyl nonane, dodecane, limone, squalene, styrene, terpinolene, β‐myrcene, (−)‐β‐caryophyllene, 1‐iridescence, nonene, 11‐methyl dodecanol, 2,3‐butanediol, cyclododecanol, dihydro linalool, eucalyptol, 2‐phenylethanoid, and 2‐methylheptan‐3‐one (Fig. 3(b)). Furthermore, 63 volatile components were identified in the 2:1 sample. A heat map depicting the content of each volatile compound, calculated using Eqn (2) and normalization, is presented in Fig. 3(c). Fatty alcohols were detected in the samples, consistent with findings by Huang et al. ⁴⁰ Terpene alcohols, including eucalyptol, linalool, and α‐pinitol, were also identified. Additionally, fatty alcohols such as 1‐hexanol, 2‐methyl‐butanol, and cyclododecanol have been discovered in various chicken samples. ⁴¹ , ⁴² These compounds can contribute fresh and fruity aromas to the samples and impart distinctive botanical aromas. ⁴³

Ketones, produced through lipid oxidation, the Maillard reaction, and amino acid degradation, significantly influence the aroma profiles of food products. ⁴⁰ , ⁴⁴ The 3:0 and 2:1 samples exhibited higher ketone concentrations than the control sample. Notably, acetophenone and acetone were detected in all samples. They potentially contributed minty and burnt aromatic notes. ⁴⁵ Furthermore, 2,3‐butanedione, characterized by its creamy aroma and low odor threshold, markedly impacts the overall aroma profile. ⁴⁶ This compound was identified in mixed bacteria samples and those containing L. sakei but was absent in the control and S. xylosus samples. Additionally, eight distinct esters were detected across various chicken samples, imparting fruity and spicy aroma notes to the product. ⁴⁷

Furans are produced through Maillard reactions and oxidative degradation processes during thermal treatment or exposure to ionizing radiation. ²⁰ A notable example is 2‐pentyl furan, which imparts a grassy aroma and liquorice‐like aroma and is formed through fatty acid oxidation. In this study, one furan compound was identified across all samples. The compound demonstrated a significant increase in concentration following fermentation compared with the control. Given its low odor threshold, this furan compound significantly contributed to the overall aroma profile of the samples. ⁴⁸

A total of ten terpene compounds were detected across all samples. These terpene compounds in food are synthesized via the mevalonate pathway and the 2‐C‐methyl‐d‐erythritol‐4‐phosphate pathway. ⁴⁹ Notably, the highest concentrations of terpene compounds were observed in the control and 2:1 samples, while the lowest levels were identified in the control group. The volatile characteristics of food play a crucial role in determining its aroma profile.

The results demonstrated distinct variations in the aroma profile of chicken breast inoculated with different inoculation proportions. Limonene, which imparts a lemon aroma to the product, was detected in all tested samples. Notably, the 2:1 sample exhibited a higher concentration of limonene. This compound is known for its antimicrobial and antioxidant properties, contributing to extended shelf life. ⁵⁰

Variations in aldehyde content were observed across all samples. The analysis identified six, seven, nine, six, eight, and seven distinct aldehydes in the control group and in samples with inoculation proportions of 3:0, 2:1, 1:1, 1:2, and 0:3, respectively.

Benzaldehyde and heptanal were detected in all samples analyzed. Aldehydes in food products primarily originate from the oxidation of spices and unsaturated fatty acids. ⁵¹ These compounds significantly influence chicken aroma owing to their low odor threshold and high volatility. One study ⁵² indicated that fatty aldehydes, including butyraldehyde, glutaraldehyde, hexanal, heptanal, and nonanal, played a crucial role in enhancing the aroma profile of chicken products. These aldehydes contribute to the sweet and fruity aroma, fresh scent, and distinctive anise‐like aroma characteristic of chicken.

Free amino acid analysis

Figure 4(a) presents a heat map illustrating the free amino acid composition across different treatments of chicken breasts. The analysis identified 17 free amino acids in the fermented chicken breast, categorized as follows: five fresh‐tasting amino acids (Asp, Glu, Gly, Ala, Lys), four sweet‐tasting amino acids (Thr, Ser, His, Pro), two aromatic amino acids (Tyr, Phe), five bitter‐tasting amino acids (Val, Met, Ile, Leu, Arg), and one sulfur‐containing amino acid (Cys). Free amino acids serve as taste precursors, and their relative proportions significantly influence meat taste. The total free amino acids in all fermented samples ranged from 12.49% to 16.22%, suggesting that various inoculation proportions facilitate Maillard reactions and Strecker degradation. These processes involve carbonyl compounds undergoing transamination reactions with amino acids, amino acid decarboxylation to aldehydes, and amino acid condensation reactions with sugars, resulting in diverse flavor profiles in the chicken. The highest concentration of free amino acids in fermented chicken breast (reaching 159% of that in the control group) was observed in the 2:1 sample. Furthermore, the analysis revealed that Glu, Lys, and Ala were the predominant free amino acids in the chicken samples.

Figure 4.

Amino acid content of differently treated chicken breasts: (a) heat map of free amino acids; (b) flavor‐presenting amino acids.

Figure 4(b) illustrates the heat map of taste‐active amino acids. According to Eqn (3), a TAV greater than 1 (TAV > 1) indicates a significant contribution to the overall taste. ⁵³ Six free amino acids (Glu, Arg, His, Lys, Ala, and Pro) were identified as the main taste‐active amino acids, as their TAV values exceeded 1. The highest aroma intensity of Glu was 8.093–20.747 in the 1:2 group, indicating that Glu was the key taste‐active amino acid in fermented chicken breast and contributed markedly to the umami and sweet taste profile. Glu and Ala are also primary sweet amino acids, which have the effect of reducing bitterness, enhancing freshness, and improving palatability. ⁵⁴ Furthermore, higher levels of Glu and Ala confer a high degree of freshness and sweetness to chicken breasts. ⁵⁵ The aroma intensity of Arg ranged from 2.9 to 14.44. Arg, a bitter and semi‐essential amino acid, increases the complexity of aroma profile and enhances freshness.

CONCLUSION

This study demonstrated that the optimal fermentation conditions for chicken breast were an L. sakei M2/S. xylosus Y4 ratio of 2:1, a total inoculum quantity of 1 × 10⁷ CFU g⁻¹, fermentation at 30 °C for 24 h, and a final pH of 5.12. The fermentation process enhanced elasticity and chewiness while reducing hardness, decreased moisture content and water activity, increased the a* value, significantly elevated the content of taste‐active amino acid, and increased the abundance of volatile aroma compounds. The fermentation treatment combining L. sakei and S. xylosus markedly improved chicken breast meat quality, suggesting potential commercial applications. However, the interactions between the two fermentation microorganisms, as well as their interactions with raw materials and additives during the fermentation process, remain to be elucidated. Future research could focus on in‐depth investigations of these microbial interactions and their roles in meat products. Such studies could lead to further optimization of production processes and product quality, potentially resulting in the development of novel, nutritious, and healthy fermented meat products with unique aromas and improved textures.

AUTHOR CONTRIBUTIONS

Yajun Liu: data curation, formal analysis, investigation, writing – original draft. Shengxiang Qi: conceptualization, methodology, software, supervision, writing – review and editing. Hui Zhou: methodology, investigation, writing – review and editing. Kezhou Cai: methodology, investigation, writing – review and editing. Xianbao Li: methodology, software, writing – review and editing. Zongyuan Zhen: conceptualization, methodology, formal analysis, validation, visualization, writing – review and editing. Jingjun Li: conceptualization, methodology, resources, writing – review and editing, supervision, funding acquisition, project administration.

CONFLICT OF INTEREST

The authors declare no conflicts of interest.

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available from the corresponding author upon reasonable request.