Flavor formation analysis based on sensory profiles and lipidomics of unrinsed mixed sturgeon surimi gels

Highlights

• •The compound surimi was prepared with 40% chicken breast addition.
• •Overall taste profile and 72 volatile compounds were identified.
• •A number of 1594 lipid molecules were determined by non-targeted lipidomics.
• •The flavor formation was contributed by glycerol phospholipid oxidation.

Abstract

New insights revealing the flavor formation of unrinsed mixed sturgeon surimi with chicken breast were evaluated. Seventy-two volatile compounds were identified by gas chromatography–ion mobility spectrometry among the 11 surimi sample groups. The addition of 40% chicken breast caused changes in the concentrations of amino acids, 5′-nucleotides, and organic acids. Sensory attributes of balsamic, waxy, green, fresh, fatty, citrus, and aldehydic were marked when corelated with 125 volatiles identified by comprehensive two-dimensional gas chromatography–time-of-flight mass spectrometry. A total of 357 different lipids were identified through UPLC-Q-Orbitrap. Analysis of the correlations between flavor-active compounds and 16 different lipids revealed that various pathways, including the degradation of triglycerides, the biosynthesis of phosphatidylcholine, and the biosynthesis of lysine, serine, and methionine, were associated with flavor formation. This study provides a theoretical basis for the development of sturgeon processing industry and surimi products from the perspective of lipid changes.

Introduction

Surimi is a wet-concentrated myofibrillar protein product obtained by continuously beating and shredding fish meat. In recent years, surimi products have gradually attracted attention in China because of the high protein and low fat (Zhao et al., 2022). The global marine fish production has declined because of the pollution of marine environment, therefore cultured freshwater fish is regarded as a potential material for surimi products (Xiao et al., 2021). The sturgeon is one of the longest-lived fish worldwide and has high economic value. Sturgeon meat was the main byproducts which accounts for 40 % of fish weight, which has become a suitable raw material for preparing surimi (Falahatkar, 2018).

Rinsing is an important step in traditional surimi processing which removes some of the pigments, enzymes, lipids, and sarcoplasmic proteins in surimi but also causes the loss of nutrients and increases the cost of industrial production (Wang et al., 2019). Non-rinsing treatment was employed for surimi, but unrinsed surimi has a high content of endogenous enzymes and lipids including triglycerides, phospholipids, and sterols, which are important active substances in sturgeon meat. These lipids, especially those containing unsaturated fatty acid chains, are easily oxidized during processing (Li, Al-Dalali, Zhou, Wang, & Xu, 2021). Among these lipids, triglycerides and phospholipids have a great impact on the flavor of meat products, mainly because they contain a large amount of unsaturated fatty acid groups (Gandemer, 2002). Phospholipids have a greater impact on flavor than triglycerides because they contain polyunsaturated fatty acid groups and are more miscible with aqueous solutions of Maillard reactants (Farmer & Mottram, 1990). Therefore, sturgeon surimi is prone to lipid oxidation during processing and storage, which results in the generation of volatile flavor compounds that impart a fishy smell and seriously impacts on the quality of surimi products (Zhu et al., 2019). It is necessary to develop a new type of surimi product in which the gelatinous quality of unrinsed surimi is increased by the addition of selected substances and the fishy smell caused by lipid oxidation is alleviated.

Chicken breast meat is considered to be a suitable raw material for improving the characteristics of sturgeon products because its whiteness is similar to that of sturgeon meat, and also can reduce cost and improve gel properties. We have demonstrated that adding chicken breast meat into unwashed sturgeon surimi has the advantages of rich in myofibrillar proteins, good gel properties, low fat content and low price (Wang et al., 2019). The chicken possesses good thermal gelling property because of the myofibrillar protein, which accounts for more than 40 % of the total chicken protein. In chicken gel products, the increase of protein content contributes to hydrogen bond, covalent bond, disulfide bond, hydrophobic interaction, electrostatic force, and ionic strength of solution to form an orderly network structure. These factors play an important role in stabilizing food quality, usually related to the palatability and flavor of the final product (Xia et al., 2022). In addition, studies confirmed that the content of umami peptides had increased in chicken meat with heat treatment, including phenylalanine, tryptophan and creatinine. Therefore, chicken can be used as an effective addition material of thermal gel and affect the flavor profile of gel products (Yu, Wang, Yin, Ge, &Liao, 2021). However, a more thorough understanding of the flavor improvement during surimi processing is still unknown.

In recent years, most studies have focused on the gel properties of surimi, including compound surimi. Liang, Zhu, Ye, Jiang, Lin, & Liu (2020) used ultrasonic-assisted treatment and microwave treatment combined with water bath heating to investigate the gel characteristics of surimi mixed with crabmeat. Jiao et al. (2019) examined the effect of the addition of fish oil on the results of microwave heating combined with conductive heating of silver carp surimi gel. Research on flavor has mainly focused on examining the influence of microbial fermentation on the flavor of surimi and correlations between proteomics data and the flavor of surimi produced by different processing methods (Li et al., 2022, Liu et al., 2022). Several studies have provided scientific evidence of the impact of lipid and protein oxidation on the loss of quality including the flavor formation during storage of a processed chicken product (Silva et al., 2018). “Grilled chicken” and “beef charqui” flavors, which were found to be linked to lipid oxidation, decreased over time in favor of rancid aroma. Yao et al. (2022) investigated the changes in flavor during the processing of Dezhou braised chicken and found that 1-hexanol, 4-methyl-2-pentanone, and 1-pentanol were mainly produced during the frying stage, and they speculated that these compounds were correlated with the oxidation of fats. Hexanal, octanal and nonanal were identified as lipid oxidation products, and may all be used as indicators of the development of “press-over-flavor” in high-pressure processed chicken breast (Wiggers, Kroger-Ohlsen, & Skibsted, 2004). At present, research on the flavor improvement of the mechanism of lipid oxidation in compound surimi comprising sturgeon with added chicken breast meat is still relatively rare.

Lipidomics can conduct a comprehensive study of lipids, including the regulation of lipid oxidation, food composition analysis, quality identification, and traceability in food research (Li et al., 2020). Compared with NMR and GC–MS, LC–MS provides abundant molecular identification through simultaneous quantification (Chen et al., 2017). However, there are relatively few reports on the flavor formation of sturgeon surimi in the lipid oxidation stage of the thermal gel process.

The purpose of this study was: (i) to determine the optimal ratio of chicken breast meat to sturgeon surimi via electronic sensory evaluation and gas chromatography–ion mobility spectrometry (GC-IMS); (ii) to identify the flavor compounds and their changes during processing; and (iii) to investigate the mechanism of flavor improvement with regard to lipid oxidation in compound chicken breast meat–sturgeon surimi via lipidomics. This study will provide theoretical guidance for the future development of high-quality, low-cost, and diverse sturgeon surimi products.

Materials and methods

Materials

Juvenile hybrid sturgeon (Acipenser baerii Brandt ♀ × Acipenser schrenckii Brandt ♂) with a weight of 1.5 ± 0.5 kg and a length of 65 ± 10 cm were purchased from the Chengyang aquatic farm (Qingdao, Shandong, China), kept in water to keep them alive, and transported to the laboratory within 1 h. Chicken breast meat was purchased from the local market in Qingdao (China). The reagents used were analytically pure and were purchased from Shanghai Yuanye Biological Technology Co., Ltd (Shanghai, China).

Preparation of samples

To prepare sturgeon surimi, live sturgeon was stunned by a blast to the head, and meat was obtained by removing the head, viscera, and skin. Then the back meat of the sturgeon was collected manually, washed to remove the blood, and finally minced using a meat grinder (Model UM5, Stephan Machinery GmbH, Hameln, Germany). A cryoprotectant (4 % sorbitol and 0.25 % sodium tripolyphosphate) was then added to the unrinsed sturgeon surimi and mixed evenly by a silent cutter (ZB-125, Zhucheng Ruiheng Machinery Factory, Shandong, China). The sturgeon surimi was packaged and frozen at −60 ± 2 °C until used. The method used for the preparation of minced chicken breast was similar to that used for sturgeon surimi. The prepared sturgeon surimi and chicken breast were combined in different proportions to prepare compound surimi products.

Samples of the abovementioned products were divided into three groups (SX: unrinsed sturgeon surimi; SJ: minced chicken breast; and SF: unrinsed sturgeon surimi with different proportions of chicken breast [w/w]). Different proportions of chicken breast (10 %∼100 %) were then added to the unrinsed sturgeon surimi. The mixtures were chopped in a mixer for 2 min, followed by additional chopping with 2.0 % salt (w/w) for 2 min. After that, the resulting paste was firmly sealed into vacuum bags before treatment at 90 °C for 30 min. All samples (X: unrinsed sturgeon surimi gels after heating; J: minced chicken breast gels after heating; and F: unrinsed sturgeon surimi gels with different proportions of chicken breast after heating[w/w]) were immediately cooled in ice water for 10 min and stored at −60 °C until further analysis.

Sensory evaluation to determine the proportion of chicken breast

Twelve panel members (six males and six females, aged 20–30) from Ocean University of China participated in the evaluation based on Sensory analysis—Methodology—General guidance for establishing a sensory profile (GB/T 39625-2020, China National Standard). Panelists were master’s and doctoral students of Aquatic Product Flavor Chemistry Laboratory of Ocean University of China, and had previously been selected based on their sensory acuity, discriminating ability. These panelists were highly experienced in sensory descriptive analysis and scores evaluation. The panelists were healthy nonsmokers with no taste/smell disorders and were recruited for the entire sensory experiment. Sensory evaluators were informed consent to participate in the sensory tests in this study, team members all participated in more than one sensory evaluation with minimal relevant experience.

The panelists were trained to evaluate the following standard taste solutions: 1 mM quinine sulfate (bitter taste), 8 mM monosodium glutamate (MSG) (umami taste), 5 mM citric acid (sour taste), and 10 mM glutathione (kokumi taste) (Liu, Zhu, Wang, Zhou, Chen, & Liu, 2020). For each sample, the chewing time in the oral cavity was not less than 3 s, and the sniffing time was not less than 5 s. To ensure the accuracy of the results, the panelists were asked to take a 2 min break and rinse their mouths with ultrapure water between each sample. A 25-point scale (0–9, unacceptable; 10–14, barely acceptable; 15–20, basically acceptable; and 21–25, completely acceptable) was used to score the taste, smell, state, and color of the samples (Sensory analysis-Guidelines for the use of quantitative response scales, GB/T 39501-2020/ISO 4121:2003) (Zhang, Zhao, Su, & Lin, 2019). Panelists were instructed to perform evaluations in the same place each time with no distractions.

All sensory experiments were performed in sensory booths under artificial lights at a temperature of 25 ± 2 °C and a humidity of 60 %. All samples were packed in odorless plastic cups (25 mL; Lvcang Co., Shanghai, China) labeled with three random codes. All products were commercially purchased or produced from commercially available food-grade materials in food-grade environments using no novel ingredients. Testing procedures followed sensory research protocols including ISO 11136 Standard of International Organization for Standardization.

Determination of flavor profiles of surimi samples by GC-IMS

A 3 g sample was placed in a 20 mL headspace sample bottle, and enrichment and equilibration were allowed to occur for 20 min at a temperature of 60 °C and an oscillation frequency of 13 g. A 500 μL sample was withdrawn from the headspace using a sample needle and injected into the injection port of a fused silica capillary column (FS-SE-54-CB, 15 m × 0.53 mm × 0.50 μm). The volatile components were then separated and injected into an ion mobility spectrometer, in which the temperature was 45 °C and high-purity nitrogen was used as the carrier gas. An LAV (version 2.21) workstation (G. A. S. Company, Dortmund, Germany) was used for identification, processing experimental data, and drawing topographic maps and fingerprints. A Library Search (version 1.0.8) workstation was used in combination with an IMS library to analyze the detected compounds. The retention index (RI) of volatile compounds was calculated using C4–9n-ketones as an external standard reference. Volatile compounds were identified by comparing the drift times and RI values with those in a GC-IMS library. Each group of samples was tested in parallel three times.

Analysis of free amino acids

An amino acid analyzer (L-8900; Hitachi Ltd, Tokyo, Japan) was used to determine the contents of free amino acids. According to the modified method, a sample (2.0 g) was extracted with 15 mL of 5 % trichloroacetic acid for three times, ultrasound for 15 min, and then allowed to stand for 2 h. The extract was centrifuged at 21,242g for 15 min. The supernatant was dilute with trichloroacetic acid to volume of 10 mL and the pH was adjusted to 2.0 with 6 M sodium hydroxide. Finally, the supernatant was filtered through a 0.22 μm microporous membrane and then injected into the amino acid analyzer Manninen, Rotola-Pukkila, Aisala, Hopia, and Laaksonen (2018).

Analysis of taste-active nucleotides

With reference to the method devised by Jiao et al. (2019) with slight modifications, 20 mL of 5 % trichloroacetic acid was added to each sample (3 g), which was homogenized for 30 s at 4 °C, and this procedure was repeated three times. The supernatant was collected after centrifugation at 21,242g at 4 °C for 15 min by a refrigerated centrifuge (LG10-3A, Beijing Medical Centrifuge factory, Beijing, China) and the abovementioned operation was repeated. A potassium hydroxide solution was added to the supernatant to adjust the pH to 5.8, and the supernatant was tested after passing through a 0.22 μm aqueous-phase filter. A high-performance liquid chromatography (HPLC) system (UltiMate 3000; Thermo Fisher Scientific, Waltham, MA, USA) was used with a C18 column (5 μm, 4.6 mm id × 250 mm, Agilent Technologies, California PA, USA). The mobile phase comprised 98 % NaH₂PO₄ (0.05 mol/L, pH 6.8) and 2 % methanol, with a detection wavelength of 254 nm, a flow rate of 0.6 mL/min, and a column temperature of 30 °C. Standards including adenosine monophosphate (5′-AMP) (≥98 %), guanosine monophosphate (5′-GMP) (≥98 %), and inosine monophosphate (5′-IMP) (≥98 %) (Shanghai Yuanye Biological Technology Co., Ltd, Shanghai, China) were used to calculate the concentrations of the respective compounds.

Analysis of organic acids

The contents of organic acids were determined by HPLC system. A 5.00 g sample was added to 25 mL ultrapure water, and centrifuged (5310 g, 4 °C, 20 min). The content of the supernatant was determined by HPLC after passing through a 0.45 μm aqueous-phase filter membrane. The HPLC instrument and chromatography column were the same as those used for the determination of nucleotides. The mobile phase comprised A: methanol and B: 0.05 % (v/v) phosphoric acid, and the flow rate was 0.6 mL/min. The detection wavelength was 210 nm. Measurements were performed three times in parallel qualitatively and quantitatively by comparison with the retention times and peak areas of standards.

Calculation of equivalent umami concentration

The equivalent umami concentration (EUC) (g MSG/100 g) refers to the MSG concentration corresponding to the umami intensity produced by the synergistic effect of umami-active amino acids and 5′-nucleotides. The formula used for calculating the EUC is as follows:

$$\text{Y}=\sum {\text{a}}_{\text{i}}{\text{b}}_{\text{i}}+12.18\sum {\text{a}}_{\text{i}}{\text{b}}_{\text{i}}\sum {\text{a}}_{\text{j}}{\text{b}}_{\text{j}}$$ (1)

where Y is the EUC value of the sample mixture in g MSG/100 g; a_i is the concentration (g/100 g) of each umami-active amino acid (aspartic acid or glutamic acid); a_j is the concentration (g/100 g) of each umami-active 5′-nucleotide (5′-IMP, 5′-GMP, 5′-xanthosine monophosphate [5′-XMP], or 5′-AMP); b_i is the relative umami concentration of each umami-active amino acid with respect to MSG (aspartic acid = 0.077, glutamic acid = 1); b_j is the relative umami concentration of each umami-active 5′-nucleotide (5′-GMP = 2.3, 5′-IMP = 1, 5′-XMP = 0.61, 5′-AMP = 0.18); and 12.18 is the synergy constant based on the concentration used (g/100 g).

Electronic tongue analysis

Electronic tongue analysis was performed using an SA-402B taste sensing system (Intelligent Sensor Technology, Inc., Kanagawa, Japan) equipped with an autosampler, a reference electrode, and a multichannel lipid/polymer membrane electrode. A sample (20 g) was thoroughly ground, and odor-active substances were extracted with 100 mL distilled water and centrifuged at 13,595 g for 15 min. The supernatant was filtered. The potential differences between sensors coated with each individual compound and the Ag/AgCl reference electrode were measured in equilibrium. Five replicate experiments were performed for each sample.

Determination of volatile compounds by two-dimensional GC–quadrupole time-of-flight mass spectrometry

The volatile compounds present in six samples were analyzed by headspace solid-phase microextraction combined with two-dimensional GC–time-of-flight mass spectrometry (GC × GC-TOFMS). A freeze-dried sample (5 g) was weighed into a 20 mL headspace vial, to which 1.5 mL ultrapure water, 0.1 mL internal standard (2,4,6-trimethylpyridine, 7.49 mol/mL), and a rotor were added, and the vial was sealed immediately. The sample was equilibrated and extracted for 30 min at 53 g at 60 °C. The fiber was then desorbed in the inlet of a gas chromatograph for detection and analysis. The GC × GC-TOFMS system comprised a gas chromatograph (8890, Agilent, California PA, USA) and a time-of-flight mass spectrometer (7250A, Agilent, California PA, USA) and was equipped with a one-dimensional column (DB-5MS, 0.25 mm × 30 mm × 0.25 mm, J&W Scientific, USA) and a two-dimensional column (DB-17MS, 2 μm × 0.18 μm × 0.18 μm, J&W Scientific, USA). The GC conditions were as follows: the injector temperature was 250 ℃, and the injection mode was splitless. The flow rate of the carrier gas (He, 99.99 % purity) was 1.2 mL/min. The initial temperature was 40 °C, which was increased to 100 °C at a rate of 5 °C/min and then increased further to 180 °C at a rate of 2 °C/min. Finally, the temperature was increased to 240 °C at a rate of 5 °C/min and held for 5 min. The time-of-flight mass spectrometry (TOFMS) conditions were as follows: the ionization energy was 70 eV with an ionization temperature of 220 °C and an interface temperature of 280 °C. The acquisition voltage was 1700 V, and mass spectrometry peaks were recorded in the range from 20 to 400 m/z at an acquisition rate of 100 spectra/s.

Lipidomics analysis

The method was a modified version of Folch’s chloroform/methanol extraction method (Folch et al., 1957). In brief, 500 mg of each sample was weighed out, and 9 mL of chloroform–methanol solution (2:1, v/v) was added. The sample was added 3 mL ultrapure water and vortexed for 1 min, allowed to stand at 4 °C for 1 h, and centrifuged using a refrigerated centrifuge at 4 °C and 5310 g for 10 min. The supernatant was extracted in a clean test tube, and the lower layer was further extracted with 6 mL of a chloroform–methanol solution (2:1, v/v). The mixed extract was gently blown dry with nitrogen and finally reconstituted with 1 mL isopropanol/acetonitrile (2:1, v/v). The sample was passed through a 0.22 μm organic filter membrane. Quality control (QC) samples were prepared by mixing equal volumes of the extracts of all samples. Liquid chromatography coupled with quadrupole orbital ion trap mass spectrometry (UPLC-Q-Orbitrap, Thermo, Massachusetts Waltham, USA) was used for lipidomics analysis. The column was a Hypersil Gold C18 reverse-phase chromatography column (100 mm × 2.1 mm × 1.9 μm); mobile phase A comprised 0.1 % formic acid + 10 mmol/L ammonium formate + acetonitrile/water (60:40, v/v); and mobile phase B comprised 0.1 % formic acid + 10 mmol/L ammonium formate + isopropanol/acetonitrile (90:10, v/v). The injection volume was 2 μL, the injection speed was 300 μL/min, and the column temperature was 55 ℃. The gradient elution program was as follows: 0–1 min, 30 %–38 % B; 1–4 min, 38 %–56 % B; 4–14 min, 56 %–98 % B; 14–15 min, 98 %–100 % B; 15–16 min, 100 % B; 16–16.1 min, 100 %–30 % B; and 16.1–20 min, 30 % B. The electrospray ion source was employed in positive and negative ion mode. The spray voltage was 3.5 kV; the sheath gas flow rate was 48 L/min; the auxiliary gas flow rate was 11 L/min; the capillary temperature was 240 °C; the auxiliary gas heater temperature was 300 °C; the acquisition time was 0–20 min; the resolution in scanning mode was 70,000 full widths at half maximum (FWHM); the resolution in secondary scanning mode was 35,000 FWHM; and the mass acquisition range was 100–1500 m/z.

Statistical analysis

All experiments were performed independently at least three times. The results for the flavor component indicators were all based on the mean ± standard deviation. The data were analyzed by SPSS 25 software, and a value of p < 0.05 was considered to indicate a significant difference. The data were imported into the MetaboAnalyst 4.0 (https://www.metaboanalyst.ca) platform for partial least squares discriminant analysis (PLS-DA), PCA, and variable importance in projection (VIP) analysis.

Results and discussion

Sensory evaluation and GC-IMS determination of different surimi samples

Sensory evaluation was performed on 11 groups of samples of compound surimi with different proportions of added chicken breast (0 %∼100 %). The sensory scores were highest when the proportion was 40 % (Fig. 1A). However, if a higher amount of chicken breast is added, the hardness of the compound surimi will increase, its elasticity will decline after thermal gelation, and the unique taste and smell of sturgeon will gradually be lost. It can be seen from the GC-IMS fingerprint (Fig. 1B) that 72 volatile compounds were detected in the 11 sample groups. These belonged to seven categories and consisted of 24 alcohols, 7 ketones, 13 aldehydes, 4 alkanes, 5 alkenes, 6 esters, and 13 other compounds. With an increase in the amount of chicken breast meat, the contents of volatile compounds significantly changed. For example, 3-octanone has a fruity scent, 1-octen-3-ol has a scent of mushrooms and lavender, and heptanal has a fruity scent. These compounds began to appear and gradually increased in content when the proportion of chicken breast reached 40 %. Previous study also reported that the important aroma characteristics of chicken products were composed of butyraldehyde, valeraldehyde, hexanal, nonanal, nonadienal, 1-octene-3-ol, decanal, and decadienal, which was main resulted from fat oxidation (Sabikun, Bakhsh, Rahman, Hwang, & Joo, 2021). (E, E)-2,4-heptadienal has a fatty taste, cyclohexanone has an earthy odor, and 3-furanmethanol has a bitter and spicy odor. Their contents decreased with an increase in the amount of chicken breast. This was consistent with the results of sensory evaluation and indicated a weakening of the earthy smell in the compound surimi. The change of flavor profile was resulted from the interaction of protein degradation products during the heating process to produce a mixture of complex compounds, such as minerals and free fatty acids (Li et al., 2022). Ethyl acetate has a slightly fruity aroma, and (Z)-3-hexenol has a strong aroma of fresh green leaves. These compounds began to appear when the chicken breast content reached 40 % but gradually disappeared after the chicken breast content reached 80 %. Hexanal, heptanal, nonanal, and octanal were identified as the sources of off-odor of sturgeon after heat treatment, while the content of these volatile compounds decreased significantly with chicken breast addition. Moreover, the addition of chicken breast produced favorable odor substances, which thereby enhanced the quality of the compound surimi. A proportion of 40 % chicken breast was therefore chosen for the preparation of compound surimi.

Fig. 1.

A. Heat map of sensory evaluation scores of sturgeon surimi with different proportions of chicken breast. B. Fingerprint of volatile compounds released by compound surimi gel prepared by gradient addition of chicken breast. C. Results of electronic tongue analysis by principal component analysis (PCA). D. Radar chart of taste attributes detected by electronic tongue. F: unrinsed sturgeon surimi gels with 40 percent of chicken breast after heating [w/w]); X: unrinsed sturgeon surimi gels after heating; J: minced chicken breast gels after heating.

Electronic tongue analysis

The taste qualities of the three samples after thermal gelation exhibited certain differences, which were divided into unrinsed sturgeon surimi gels with 40 percent of chicken breast after heating (F, w/w), unrinsed sturgeon surimi gels after heating (X), and minced chicken breast gels after heating (J). It can be seen from Fig. 1C that the cumulative contribution of principal component 1 and principal component 2 was 89.26 %. Notably, the F group (red) had closer distance with the J group (green), which was resulted from the chicken breast addition, causing the similar flavor profile of gel products. As shown in Fig. 1D, group X scored the highest in terms of astringency and bitterness and group F scored the highest in terms of saltiness, sourness, and umami. By combining the PCA results and the radar chart, it can be seen that the umami taste of sturgeon surimi after the addition of chicken breast meat was significantly improved (from 74.67 to 92.23), which was consistent with the results for the EUC. In addition, the bitterness and astringency were reduced from 58.14 to 42.73 and 18.24 to 11.36, respectively, which was consistent with the results for the contents of free amino acids and organic acids. A study reported that the umami taste played a key role in soft-boiled chicken, meanwhile, the bitterness and astringency were significantly reduced (Zhang et al., 2022). This shows that the addition of chicken breast effectively reduced the original unfavorable tastes, improved beneficial tastes such as umami, and improved the overall taste of sturgeon surimi.

Analysis of free amino acids

Table 1 shows the contents of free amino acids in the six samples, where aspartic acid and glutamic acid provided an umami taste, threonine, serine, glycine, and alanine provided sweetness, and valine, histidine, and arginine provided bitterness. The content of umami-active amino acids in group J was 9.68 times that in group X, and group F was 3.66 times that in group X, which indicated that adding chicken breast can significantly enhance umami flavor in surimi, which was consistent with the results obtained from sensory experiments. It is worth noting that because of its low taste threshold (30 mg/100 g, wet weight) and synergy with 5′-IMP, glutamic acid had a great influence on taste perception. It has been reported that even when the concentrations of certain bitter amino acids are below their taste threshold they can enhance the sweetness and taste of other amino acids and thereby improve the overall taste (Lioe, Apriyantono, Takara, Wada, & Yasuda, 2005). After chemical conversion, the arginine content in group F was significantly higher than that in group SF (p < 0.05), and arginine was not detected in group X. Although arginine is a bitter amino acid, it improves the freshness and taste of surimi and has an important influence on the overall taste, which is shown in the form of a highly positive correlation (Zhao, Schieber, & Gaenzle, 2016). Therefore, in a combination of sturgeon surimi and chicken breast meat the arginine content increases, which thereby improves the taste quality of the product. The contributions of different taste-producing substances to the taste of aquatic products can be influenced by their interactions. When aquatic products contain a large amount of alanine or glutamic acid, the sweetness produced by glycine will be suppressed to a certain extent. Glutamic acid and AMP act in the same way and reduce the contribution of arginine to the bitter taste of aquatic products. The complex interactions between these flavor substances are one of the reasons why aquatic products exhibit unique flavors (Luo et al., 2022). Therefore, the addition of chicken breast can greatly enhance the umami taste, maintain the sweetness, and reduce the bitterness of surimi.

Table 1.

The contents of free amino acids, nucleotides, and organic acids in different samples.

	Non-volatile compounds
	F	X	J	SF	SX	SJ
Free amino acids (mg/100 g)
Asp	6.80 ± 0.020c	ND	22.59 ± 0.43a	6.19 ± 0.02c	ND	20.26 ± 0.16b
Glu	10.16 ± 0.15d	4.63 ± 0.27e	22.12 ± 0.44b	13.69 ± 0.18c	3.31 ± 0.12f	40.65 ± 0.29a
Thr	9.21 ± 0.67c	4.50 ± 0.30d	19.10 ± 0.43a	7.87 ± 0.09c	4.86 ± 0.40d	17.35 ± 0.11b
Ser	9.76 ± 0.12c	6.06 ± 0.42d	21.86 ± 0.40a	9.44 ± 0.24c	6.23 ± 0.01d	20.44 ± 0.23d
Gly	13.86 ± 0.11b	12.36 ± 0.70c	20.32 ± 0.35a	13.76 ± 0.02b	12.06 ± 0.40c	19.46 ± 0.18a
Ala	23.85 ± 0.23b	17.72 ± 1.12c	40.60 ± 0.69a	23.43 ± 0.34b	17.78 ± 0.43c	38.24 ± 0.20b
Met	4.25 ± 0.04b	2.52 ± 0.16c	9.98 ± 0.21a	3.96 ± 0.06b	2.74 ± 0.13c	9.66 ± 0.41a
Val	7.71 ± 0.04c	4.76 ± 0.28d	15.32 ± 0.20a	7.01 ± 0.27c	4.66 ± 0.22d	13.89 ± 0.18b
Ile	5.86 ± 0.06c	3.71 ± 0.18d	11.52 ± 0.24a	4.91 ± 0.23c	3.53 ± 0.07d	8.91 ± 0.59b
Leu	10.17 ± 0.13c	5.87 ± 0.36d	22.26 ± 0.48a	8.54 ± 0.17c	5.50 ± 0.05d	17.49 ± 0.29b
Tyr	4.35 ± 0.13c	2.93 ± 0.01d	8.70 ± 0.16a	3.71 ± 0.25cd	3.06 ± 0.09d	6.94 ± 0.51b
Phe	4.04 ± 0.01c	1.91 ± 0.17d	10.03 ± 0.22a	3.72 ± 0.04c	1.91 ± 0.14d	8.55 ± 0.41b
Lys	9.20 ± 0.02c	6.36 ± 0.42d	15.41 ± 0.13a	9.31 ± 0.01c	7.05 ± 0.10d	16.16 ± 0.20b
His	14.10 ± 0.29a	12.95 ± 1.15ab	12.74 ± 0.29ab	13.49 ± 0.02ab	14.01 ± 0.14a	11.95 ± 0.07b
Arg	7.22 ± 0.02c	ND	12.49 ± 0.19a	6.55 ± 0.06d	4.54 ± 0.20e	11.55 ± 0.24b
Nucleotides (mg/100 g)
GMP	5.20 ± 1.14c	4.17 ± 0.04c	7.55 ± 0.45b	7.87 ± 0.37b	5.32 ± 0.48c	13.21 ± 0.38a
IMP	34.64 ± 1.66c	46.94 ± 0.85b	24.52 ± 1.74d	69.82 ± 4.51a	74.30 ± 0.68a	45.37 ± 0.30b
AMP	7.15 ± 0.50c	6.77 ± 0.16c	10.45 ± 0.61b	9.49 ± 0.32b	7.68 ± 0.13c	13.59 ± 0.88a
EUC (g/100 g)	1.94 ± 0.06b	0.84 ± 0.05d	3.55 ± 0.03a	1.36 ± 0.06c	0.39 ± 0.01e	3.54 ± 0.19a
Organic acids(mg/g)
Tartaric acid	567.67 ± 4.31a	615.64 ± 25.92a	576.31 ± 7.13a	593.80 ± 23.14a	555.32 ± 28.80a	6.19 ± 0.61a
Malic acid	30.83 ± 0.60c	29.63 ± 4.04c	54.93 ± 0.55a	28.20 ± 1.19c	20.20 ± 1.39b	44.24 ± 3.04b
Lactic acid	18.28 ± 0.08b	12.58 ± 0.49cd	30.45 ± 0.20a	17.14 ± 0.69bc	11.33 ± 0.55d	29.61 ± 2.93a
Succinic acid	20.70 ± 4.75bc	45.10 ± 0.89a	6.99 ± 0.99cd	16.80 ± 3.69bc	23.03 ± 3.28b	6.09 ± 0.32cd

a-e Different letters in the same row demonstrate significant difference at different approach (p < 0.05).

F: unrinsed sturgeon surimi gels with 40 percent of chicken breast after heating [w/w]); J: minced chicken breast gels after heating; SF, SX, and SJ means samples without heating.

Analysis of taste-active nucleotides

Nucleotides are among the main umami-active substances in aquatic products. The main representative umami-active nucleotides are 5′-AMP, 5′-IMP, and 5′-GMP. Table 1 shows the contents of flavor-active nucleotides and the equivalent MSG concentrations (EUC values) in the six samples. It can be seen from Table 1 that the contents of 5′-GMP and 5′-AMP in sturgeon surimi were lower than those in chicken breast but the content of 5′-IMP in sturgeon surimi was higher than that in chicken breast (p < 0.05). Because 5′-IMP easily decomposes when heated, in comparison with the unheated gel groups (SF, SX, and SJ), the content of 5′-IMP after the gel was heated was significantly reduced. The degradation of 5′-IMP will produce bitter odor-active substances, the preparation of compound surimi can effectively reduce the content of 5′-IMP degradation products and thus alleviate possible bitterness. It is worth noting that 5′-AMP suppresses bitter tastes and can thus endow food with ideal sweetness and umami taste. The taste characteristics of 5′-IMP are similar to those of 5′-AMP in that it can enhance umami and sweet tastes of foods and has a synergistic effect with sweet amino acids (Chen & Zhang, 2007). After the production of compound surimi, the content of 5′-AMP was increased, which was consistent with the results of sensory evaluation. The EUC is used to reflect the synergistic effect of umami-active free amino acids and flavor-active nucleotides and is currently widely used to determine the intensity of the umami flavor of foods. It can be seen from Table 1 that the EUC values of the three samples after thermal gelation significantly increased, but the EUC value of sturgeon surimi was always less than 1 g/100 g. The EUC value of the compound surimi reached 1.94 g/100 g, which was 2.31 times that of sturgeon surimi, and the EUC value of chicken breast was 3.55 g/100 g, which showed that adding chicken breast can effectively enhance the umami taste of sturgeon surimi while reducing its bitterness and distinctive unfavorable smells.

Analysis of organic acids

According to some reports, lactic acid and succinic acid are the main metabolites in certain animal muscles, and both play an important role in the umami characteristics of aquatic products (Kani, Yoshikawa, Okada, & Abe, 2008). It can be seen from Table 1 that four organic acids were detected, the tartaric acid content was the highest and the lactic acid was the lowest, which had their unique sour taste and enable the surimi gels to produce a special “umami taste”. According to relevant research, tartaric acid is mostly found in plants, while sturgeon is herbivorous fish and infiltrates into fish body under the influence of algae in water, leading to an increase in concentration (Mouritsen, Duelund, Calleja, & Frøst, 2017). After thermal gelation, the content of tartaric acid decreased and the lactic acid content increased, which indicated that high temperatures caused a reduction in the tartaric acid content, which had a positive effect on the contents of the other organic acids. This result was consistent with the previous study, which proved the decrease of the tartaric acid content and increase of the lactic acid content for surimi gel with high temperature treatment (Luo et al., 2022). The reason is that tartaric acid participated in the Strecker reaction of proteins during high temperature treatment and degraded into furans and pyrazines. A study reported that lactic acid was an organic acid in aquatic products, suggesting that the formation of flavor in the reheated samples was related to the increase of lactic acid content (Gao, Zheng, Zhou, Tian, & Yuan, 2019). Succinic acid is a characteristic flavor component in aquatic products and provides umami and sour tastes. The content of succinic acid in sturgeon surimi was 7.5 times that in chicken breast and 2.2 times that in compound surimi, which showed that the characteristic organic acids in sturgeon meat can be retained to a great extent after the preparation of compound surimi, and succinic acid can improve the taste of sturgeon surimi in combination with other organic acids. Lactic acid can improve the buffering capacity and enhance the taste of foods. The results show that the addition of chicken breast increased the lactic acid content of sturgeon surimi, which indicated that it had a positive effect on the taste of the product. In general, the organic acid content in sturgeon gel increased after adding chicken, indicating that chicken addition had a positive effect on the retention and increase of flavor components in surimi gel.

Analysis of volatile flavor compounds determined by GC × GC-TOFMS

A number of 125 volatile compounds in four samples (F, X, J, and SF) were identified including 21 aldehydes, 18 ketones, 13 alcohols, 21 alkanes, 23 alkenes, and 29 other compounds (Supplementary Table 1). To visualize a large amount of data by reducing the dimensionality, OPLS-DA was performed on the results of GC × GC-TOFMS. The score chart generated by OPLS-DA is shown in Fig. 2C. The first and second principal components explained 28.7 % and 32.6 %, respectively, of the cumulative contribution to the variance.

Fig. 2.

A1–A4. Radar chart showing a pairwise comparison of the flavor profiles of the samples. B. Venn diagram of volatile compounds in samples. C. Results of orthogonal partial least squares discriminant analysis (OPLS-DA) of volatile compounds in samples. D. Network diagram of types of volatile compounds and flavor attributes of samples. E. Heat map of the types and contents of aldehydes and ketones in samples. F. Sankey diagram of types of volatile compounds and flavor properties of samples. X: unrinsed sturgeon surimi gels after heating; J: minced chicken breast gels after heating; F: unrinsed sturgeon surimi gels with 40 percent of chicken breast after heating [w/w]); SF: unrinsed sturgeon surimi with 40 percent of chicken breast [w/w]).

The correlation between sensory attributes and volatiles profile were identified and shown in Radar charts (Fig. 2A1–A4). Balsamic, balsam, waxy, green, fresh, fatty, fat, citrus, aldehydic, and alkane were marked. Fig. 2D and 2F show the correlation between the types of volatile compounds and typical flavor attributes of all the samples. Decane, heptanal, hexadecane, nonadecane, nonanal, styrene, and tetradecane were the main dominant volatiles among the four samples. An analysis of the types of volatile compound among the samples (Fig. 2B) found that nonanal, heptanal, and 2,5-dihydroxybenzaldehyde were the aldehydes and 4-octanone was the ketone that were common to the samples. Nonanal has aromas of roses and citrus; heptanal has a fruity aroma accompanied by a strong oily smell; and 2,5-dihydroxybenzaldehyde has aromas of bitter almond, cherry, and nut. Owing to the low flavor thresholds of aldehydes and ketones, they play a very important role in the formation of flavor (Fig. 2E). Generally speaking, saturated linear aldehydes in aquatic products are usually mixed with volatile C8 and C9 compounds and together have an impact on the flavor of fish, and aldehydes generally have a grassy fragrance and fruity and nutty aromas at low concentrations. At high concentrations, they may produce an unpleasant fishy smell (Benet, Guàrdia, Ibañez, Solà, Arnau, & Roura, 2016). As shown in Fig. 2, the concentration of aldehydes such as nonanal, heptanal, and octanal in unrinsed sturgeon surimi gels after heating was higher than that in unrinsed sturgeon surimi gels mixed with chicken breast. These aldehydes are considered to be the main sources of fishy aromas in fish. Aldehydes may be produced via the cleavage of peroxides formed after the oxidation of unsaturated fatty acids. For example, Drumm and Spanier (1991) confirmed that hexanal is the main degradation product of omega-6 fatty acid peroxides. It is important to continue to investigate the effect of lipid oxidation on flavor production in composite surimi, for which the relationship between these processes may provide the basis. Ketones have a taste similar to those of eucalyptus and fat and have low taste thresholds. Most ketones are the products of thermal oxidation and degradation of unsaturated fatty acids and have a superimposing effect on the fishy taste of aquatic products (Li et al., 2022). For example, 2-undecanone has an oily smell and was present in sturgeon surimi but not detected in the other samples, whereas 2-nonanone has a fruity and creamy smell and was detected in compound surimi and chicken breast meat but not found in sturgeon surimi. Alcohols are mainly produced by the decomposition of secondary oxidation products of fatty acids, the action of lipoxygenase on fatty acids, the oxidative decomposition of fats, and the reduction of carbonyl compounds (Wu et al., 2021). Lipid oxidation plays a very important role in the preparation of compound surimi and thermal gelation. It masks the release of substances with fishy aromas, and also produces a desirable smell during the process. However, the specific classification of lipids and the metabolic pathways involved in the flavor formation are unknown.

Lipid compositions of different samples

A number of 1594 lipids were identified based on high-resolution non-targeted lipidomics, which belonged to 14 main categories (Fig. 3A). These included 124 (7.80 %) diradylglycerols, 6 (0.38 %) monoradylglycerols, 532 (38.38 %) triradylglycerols (TGs), 300 (18.82 %) phosphatidylcholines (PCs), 29 (1.82 %) lysophosphatidylcholines, 250 (15.68 %) phosphatidylethanolamines (PEs), 21 (1.32 %) lysophosphatidylethanolamines, 39 (2.45 %) phosphatidylserines, 24 (1.51 %) sphingomyelins, 41 (2.57 %) phosphatidylinositols (PIs), 34 (2.13 %) cardiolipins, 26 (1.63 %) phosphatidylglycerols (PGs), 54 (3.89 %) ceramides, 10 (0.63 %) monogalactosyl diglycerides, and 104 (6.52 %) other lipids. As shown in an Upset diagram of the lipid compositions of unrinsed sturgeon surimi gels with 40 percent of chicken breast after heating [w/w]) (F), unrinsed sturgeon surimi gels after heating (X) and minced chicken breast gels after heating (J) (Fig. 3B), the three samples together contained 1499 lipids, of which 26 lipids were common in F and J, and 35 lipids were common in F and X. The numbers of lipids only found in a single sample group were 6 lipids in unrinsed sturgeon surimi gels with 40 percent of chicken breast after heating, 12 lipids in unrinsed sturgeon surimi gels after heating, and 7 lipids in minced chicken breast gels after heating. The identification results revealed the huge diversity and complexity of the lipids in the six samples, as well as their chemical structures, composition, and polarity.

Fig. 3.

A. Pie chart of the types and contents of lipid molecules in samples. B. S-plot of the types of lipid molecules in samples. C. Venn diagram of the types of lipid molecules in samples. D. Graph of PCA results for samples. E. Graph of OPLS-DA results for samples. X: unrinsed sturgeon surimi gels after heating; F: unrinsed sturgeon surimi gels with 40 percent of chicken breast after heating [w/w]); J: minced chicken breast gels after heating; SF, SX, and SJ means samples without heating.

By analyzing the distribution of glycerophospholipids in the samples, it could be seen that 743 glycerophospholipids were detected in all the cooked samples, 14 glycerophospholipids were common in F and X, and 11 glycerophospholipids were common in F and J. Because glycerophospholipids are important components of cell membranes, their content in animals is higher than those of other lipids. Moreover, glycerophospholipids contain more unsaturated fatty acids, of which the oxidation will produce large amounts of aldehydes and ketones, which are the main volatile flavor compounds, and thus their contribution to flavor is much greater than those of other lipids (Ba, Amna, & Hwang, 2013). A PCA score plot was employed to visualize the separation of the six samples. As shown in Fig. 3D, clear differences were observed, with all scores located in the 95 % confidence ellipse, and the contributions to the variance explained by the first two principal components were 51.9 % and 12.5 %, respectively. As shown in Fig. 3E, the six samples were clearly classified according to the first and second principal components based on PLS-DA, and the cumulative contribution was 61.2 %. The samples obtained after the six treatment methods were separated from each other with good discrimination, and the results were consistent with those of PCA. This shows that the differences between the samples were clear and the differential lipids could be screened by their VIP values (Jia et al., 2022).

Analysis of key differential lipids in different samples

A larger VIP value and a smaller p value indicate a greater difference in the target lipids between the respective samples. According to the results (VIP ≥ 1 and p ≤ 0.05), 537 differential lipids were identified between F and J, 324 differential lipids were identified between F and X, and 182 differential lipids were identified between unrinsed sturgeon surimi gels mixed with chicken breast after heating and samples without heating. A total of 13 lipids were detected in all three groups (Fig. 4D), namely, PC(14:0e/18:2), PC(18:1e/20:4), PC(18:2/20:4), PC(18:3/22:6), PC(19:1/22:6), PC(18:4/22:6), PI(22:6/22:6), PI(20:0/20:4), TG(18:0/16:0/20:1), PG(18:1/18:2), PE(8:1e/13:0), PE(8:0e/14:4), and MePC(18:1e/18:2). In addition, TG (20:1e/16:1/18:2), PC(16:2e/22:6), PI(16:1/22:6), PC(14:0/18:2), TG(16:1/16:1/18:2), PE(14:1e/18:1), TG(18:0e/16:0/16:0), MePC(10:0/20:4), and TG (20:0e/16:0/16:0) were the lipids with the largest VIP values among the three groups. According to visual analysis of the degree of difference in the lipid content (Fig. 4A, 4B, & 4C), it can be seen that there were significant differences in lipid content between different groups. This indicated that combining sturgeon surimi and chicken breast changed the lipid composition, the types of lipid present after the preparation of compound surimi were more complex than in sturgeon surimi or chicken breast alone, and there were obvious changes in lipid content.

Fig. 4.

A-C. Differences in lipids between two groups of samples. D. S-plot of the differences in lipids between groups of samples. E. Heat map of correlations between 16 different lipids and the flavor-active compounds with the smallest p values (*p < 0.05, **p < 0.01, ***p < 0.001) in each group according to the differential lipids identified by comparison between the groups.

Correlations between key flavor compounds and lipid molecules

According to the results of pairwise comparisons of the three groups of samples, the lipid molecules with the most significant differences were selected for analysis of their correlations with taste- and odor-active substances, and the results are shown in Fig. 4E. Aspartic acid, alanine, glycine, serine, and threonine are amino acids with umami and sweetness properties. They are associated with PE(14:1e/18:1), PI(20:0/20:4), PC(16:2e/22:6), PC(14:0/18:2), TG(18:0e/16:0/16:0), TG(20:0e/16:0/16:0), TG(20:1e/16:1/18:2), and other lipid molecules, with which they had a significant positive correlation. With regard to F and J, the content of the differential lipid PC (14:0/18:2) was higher in chicken breast and was closely correlated with the content of sweet amino acids, which indicated that the content of sweet amino acids in chicken breast was higher than that in compound surimi, which was consistent with the results of the determination of free amino acids. Among taste-active nucleotides, 5′-GMP and 5′-AMP are mainly associated with PE(8:0e/14:4), PE(14:1e/18:1), TG(18:0/16:0/20:1), TG(20:0e/16:0/16:0), and TG(18:0e/16:0/16:0), with which they had a positive correlation. Studies have shown that 5′-AMP suppresses bitterness, imparts ideal sweetness and umami to foods, and in combination with sweet amino acids produces a synergistic effect (Chen & Zhang, 2007). Therefore, in compound surimi the off-flavors caused by thermal gelation of sturgeon surimi can be avoided and pleasant taste-active substances can be produced. According to the correlation analysis, lactic acid was mainly positive associated with unsaturated glycerophospholipids, such as PE (14:1e/18:1) and PC (14:0e/18:2). The addition of chicken breast compensated for the lack of lactic acid in sturgeon surimi and improved the flavor of the samples. This may be due to interactions between lipid oxidation and proteins or because lipid oxidation products can induce aggregation of protein molecules and thus cause changes in the structures and functional properties of proteins (Huang, Hua, & Qiu, 2006). As the main components of proteins, amino acids are sensitive to lipid oxidation. In particular, aromatic amino acids are very susceptible to lipid oxidation. Aliphatic amino acids are affected by lipid oxidation via the removal of hydrogen atoms from α-carbon atoms to generate carbon-centered free radicals. Other types of amino acids are also affected by lipid-induced oxidation (Soyer, Özalp, Dalmış, & Bilgin, 2010).

The predicted map of flavor formation pathways based on omics data

Researchers have found that the interaction of amino acids and lipid oxidation products will produce condensation products that affected the aroma (Adams, Kitrytė, Venskutonis, & De Kimpe, 2009). The aldol condensation reaction of carbonyl compounds among lipid oxidation products plays a very important role in the initial reactions of systems containing various metabolites formed by lipid oxidation. These compounds can continue to undergo the Maillard reaction with free amino acids or interact with intermediate products of the Maillard reaction to form volatile compounds such as aldehydes, ketones, and pyridines. As shown in Fig. 4E, nonanal, (E, E)-2,4-heptadienal, (Z)-2-heptenal, 2-undecanone, and octanal are all volatile compounds characterized by a fatty or oily smell. These compounds had positive correlations with unsaturated glycerophospholipids such as PI (22:6/22:6), PC (16:2e/22:6), and PI (16:1/22:6). This was consistent with a conclusion in the literature that linear aldehydes such as hexanal, octanal, and nonanal are mainly derived from the peroxidation of unsaturated fatty acids (Sidira, Kandylis, Kanellaki, & Kourkoutas, 2015). Lipid oxidation will occur at any temperature; regardless of whether the temperature is high or low, the pathways are similar. The main site of oxidative attack is a methylene group adjacent to the double bond, and the initial step involves removing a hydrogen atom from the methylene group (Giménez, Gómez-Guillén, Pérez-Mateos, Montero, & Márquez-Ruiz, 2011). The free radicals thus generated can undergo rearrangement before reacting with oxygen, which leads to the formation of many different hydroperoxides, which are unstable and can produce further products (Whitfield & Mottram, 1992). Numerous lipid molecules, especially glycerophospholipids, were involved in the regulation of taste- and odor-active substances. The unique flavor of compound surimi was produced as a result, and, at the same time, the original earthy smell and fishy smell of sturgeon surimi were improved.

Using the comparisons between the sample groups to generate a schematic diagram of the KEGG functional pathways, it was found that the effect of the addition of chicken breast on the compound surimi mainly involved the following processes: (i) the biosynthesis of PC; (ii) the degradation of triglycerides; and (iii) the biosynthesis of lysine, serine, and methionine. The results showed that fatty acids and amino acids are the main precursors of the flavor-active compounds in compound surimi (Fig. 5A). The appearance of these differential lipids indicated that interactions between lipid molecules occurred during the preparation of compound surimi from sturgeon surimi and chicken breast meat, in which lipid oxidation was likely to have played a key role. The addition of chicken breast significantly increased the abundances of PC(18:0/18:1), PC(18:4/22:6), TG(20:0e/16:0/16:0), and TG(18:0e/16:0/16:0) (Fig. 5B1-5B4). These lipids are to highlight the mechanisms and chemical compounds responsible for chicken meat flavor and off-flavor development (Jayasena, Ahn, & Jo, 2013). 1-Octen-3-ol is a characteristic volatile component of aquatic products and mainly exhibits earthy and mushroom flavors. It is produced by the reaction of alkoxy groups and fat molecules during lipid oxidation (Huang et al., 2022) and was positively correlated with PC(14:0/18:2) and PI(20:0/20:4). Researchers have proved that different hydroperoxides formed from unsaturated fatty acids produce a large amount of volatile products during lipid oxidation via different decomposition pathways. For example, phospholipid hydroperoxides are cleaved to form long-chain oxidation products and short-chain oxidation products. Under the action of enzymes such as lipoprotein lipases, lipolysis occurs to form free fatty acids. Finally, phospholipid molecules form alkanes, alkenes, aldehydes, ketones, alcohols, esters, and acids and thus produce volatile compounds (Tu, Qi, Shui, Lin, Benjakul, & Zhang, 2022). As mentioned above, the umami taste of sturgeon surimi was improved by the addition of chicken breast, which mainly reflected the synergistic effect of glutamic acid and 5′-IMP. The associated pathways and mechanisms of lipid oxidation, nucleotide metabolism, and organic acid metabolism still need to be further investigated and verified.

Fig. 5.

A. Schematic diagram of potential sources of representative flavor-active compounds based on a comparison of the KEGG pathways in compound surimi, sturgeon surimi, and chicken breast. B1-B4. The abundances of PC(18:0/18:1), PC(18:4/22:6), TG(20:0e/16:0/16:0), and TG(18:0e/16:0/16:0) in different samples. X: unrinsed sturgeon surimi gels after heating; J: minced chicken breast gels after heating; F: unrinsed sturgeon surimi gels with 40 percent of chicken breast after heating [w/w]); SF: unrinsed sturgeon surimi with 40 percent of chicken breast [w/w]).

Conclusions

The addition of chicken breast greatly concealed the release of substances with fishy aromas in sturgeon surimi and produced desirable flavor-active substances with fruity and creamy aromas such as 2-nonanone. After analyzing differences in the 1594 identified lipid molecules between the groups, it was found that glycerophospholipids are closely involved in the process of lipid oxidation and generate aldehydes and ketones such as nonanal and (E,E)-2,4-heptadienal. This study clarified the effect of the addition of chicken breast on the taste and smell of compound surimi from the perspective of lipidomics. However, owing to the complexity of the subject, except for the advantages of mixed sturgeon surimi including rich in myofibrillar proteins, good gel properties, low fat content and low price, more work needs to be carried out on the impact of lipid and protein oxidation on the loss of quality during storage.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

The following are the Supplementary data to this article:

Data availability

No data was used for the research described in the article.