Characterization of volatile flavor compounds from fish maw soaked in five different seasonings

Highlights • The flavor of fish maw soaked in five different seasonings was studied.• Organic sulfide and aromatic compound are the main components of seasoned fish maw.• E-nose and HS-GC-IMS can distinguish five kinds of seasoned fish maw.• The mixed seasoning group can effectively improve the flavor of fish maw.• The characteristics of volatile components in seasoned fish maw were visualized.


Introduction
Fish maw, the dried swim bladders, is filled with gas and is the organ that fish use to maintain balance and sink and float.Fish maw has been promoted and consumed in Asia as a traditional delicacy and premium tonic (Clarke, 2002).Especially in China and Southeast Asia, fish maw has been used as a luxury tonic due to its multiple medicinal properties (such as promoting postpartum recovery and healing of cutaneous wounds; enhancing immunity and memory; and removing free radicals) (Dai et al., 2020;Howaili, Mashreghi, Shahri, Kompany, & Jalal. 2020;Jian & Wu, 2003;Yu, Yin, Qian, & Yan, 2009;Zhao et al., 2016).The demand and value of fish maw have risen with the development of society and are expected to continue to grow in the future.
Fish maw is usually prepared into soup or other dishes, and is favored for its soft texture and rich nutrition (Sadovy de Mitcheson et al., 2019;Sinthusamran & Benjakul, 2015).Flavor is one of the most important indicators to measure food quality.Volatile compounds in food have a decisive influence on the flavor of food.However, there has been little scientific reporting on the flavor of fish maw compared to other seafood products.Dried fish maws are rehydrated by soaking before eating, which is the preparation work required before cooking (Clarke, 2004;Newman, 2004).Due to the high flavor threshold of volatile compounds in fish maw, seasoning should be added to optimize the flavor during processing.In order to elucidate the flavor changes of fish maw during initial processing, we studied the changes of volatile compounds in fish maws soaked in water containing different seasoning (Li, Dong, Jiang, Qi, & Lin, 2022).
In the study of food flavor, headspace solid-gas chromatography-ion mobility spectrometer (HS-GC-IMS) has been widely used in the analysis of VCs in food.HS-GC-IMS is a technique for the rapid detection of volatile components in samples using GC combined with IMS (Xu, et al., 2023), which combines the high separation of gas chromatography and the high sensitivity of ion mobility spectrometry compared to the mainstream technique (GC-MS) used to study volatile flavor components (Chen et al., 2020).GC-IMS has been widely used for the detection of volatile flavor compounds in the field of food analysis (Li et al., 2020).Previously, our group also applied HS-GC-IMS technology to carry out a series of studies on food flavor (Chen et al., 2021;Jiang et al., 2022;Jiang et al., 2022;Jin et al., 2023).
In this study, HS-GC-IMS combined with electronic nose technology were applied to clarify the evolution and formation of volatile compounds in fish maw during soaking with different seasoning (deionized water, onion, gingerroot, Sichuan pepper, mixed seasonings), and the flavor fingerprints of fish maw were also established.Furthermore, principal component analysis (PCA) and heat map analysis were used to distinguish the characteristic compounds among different treated samples.This study aimed to provide theoretical evidence to control flavor and quality changes of fish maw during soaking.The research results will provide reference for producing high-quality fish maw and improving the flavor quality of fish maw during processing.

Material
Dried fish maws (Gadus morhua) were purchased from Beijing Tong Ren Tang Health (Dalian) Seafoods Co., Ltd.Onion, ginger, Sichuan pepper purchased from the local markets.Firstly, the cleaned fish maw was boiled in a water bath at 100 • C for 8 min, and then the fish maw was divided into 6 equal portions (500 g each) and soaked in different seasoning groups (A-F) for 24 h.A: control group (the boiled fish maw was not further treated); B: deionized water group (the boiled fish maw was soaked in deionized water (3000 mL)); C: onion group (onion (300 g) was boiled in deionized water (3000 mL) for 60 min, and cooled to obtain onion seasoning water); D: ginger group (ginger (300 g) was boiled in deionized water (3000 mL) for 60 min, and cooled to obtain ginger seasoning water); E: Sichuan pepper group (Sichuan pepper (75 g) was boiled in deionized water (3000 mL) for 60 min, and cooled to obtain Sichuan pepper seasoning water); F: mixed seasonings group (onion (300 g), ginger (300 g) and Sichuan pepper (75 g) were boiled in deionized water (3000 mL) for 60 min to obtain mixed seasoning water).The soaked fish maws were then used for subsequent experiments immediately.

Sensory evaluation
A total of 12 panelists (8 females and 4 males, aged from 19 to 30) ranked the fishy smell of boiled and soaked fish maws.In order to ensure the homogeneity of the samples tested by the panelists, the samples were divided into 10 equal parts, and the samples were divided into equal parts after mixing.After packaging, the samples were marked with 3 random numbers, and the samples were handed over to panelists for sensory evaluation in orthogonal Latin order.The evaluations were conducted in the evaluation compartment of the sensory laboratory, and the evaluators were not allowed to communicate with each other.Sensory evaluators observed, smelled, and tasted the samples according to the given sample number sequence, and then ranked the products according to a certain characteristic of the products.Sensory evaluators need to clear their mouths with water and chew unsalted soda crackers during the tasting interval (30 s) to remove the residual taste (Zhang, Ayed, Wang, & Liu, 2019).The evaluators ranked the samples according to the fishy smell, the fishiest ranked 1, and the least fishy ranked 6. Statistical analysis was carried out on the sorting results by Friedman test to realize the significance analysis of differences between samples.Multiple comparisons with the least significant difference value.The sensory data were collected and calculated by the online sensory analysis system SE, and the comparison value R between the two samples was obtained.

Electronic nose (E-nose)
The volatile compounds of the six group of boiled and soaked fish maws were determined using an electronic nose (PEN3.0,AIRSENSE, Germany).Boiled and soaked fish maw sample (1 g) was placed in a glass injection bottle and then equilibrated at 25 • C for 15 min.The detection parameters were set as follows: sample preparation time (5 s), detection time (60 s), auto -zero time of 10 s, and sensor cleaning time (60 s) and period (54-56 s) (Ma, Mu, & Zhou, 2021;Weng, et al., 2021).

Statistical analysis
All experiments were performed at least in three independent trials.The data were analyzed by SPSS 18.0 software (SPSS Inc., Chicago, IL, USA) and statistical significance of differences with p < 0.05 was evaluated with LSD test.According to all volatile compounds identified between samples, a plug-in PCA score and biplot diagram (through normalization and eigenvectors) by the HS-GC-IMS instrument were also acquired.

Rank sum test statistical sorting results
The addition of seasonings can effectively improve the overall flavor of meat products (Huang et al., 2020).The results of least significant difference (LSD) analysis were shown in Table 1, The ranking results of the rank-sum test for fish maw flavor were shown in Table 2.According Table 1 LSD analysis of flavor adsorption effect of fish maw.
Note: N means there is no significant difference between samples, Y means there is a significant difference between samples (p < 0.05).A: unseasoned group, B: deionized water group, C: onion group, D: ginger group, E: Sichuan pepper group, F: mixed seasonings group.

Table 2
Sorting results of flavor adsorption effect of fish maw.Note: Different lowercase letters in the same line represent the significant test results of LSD.The same letter means no significant difference between samples, while different letters mean significant difference between samples (p < 0.05).
A, B, C, D, E, and F are the same as Table 1.
X. Miao et al.
to Table 1 and Table 2, the score of fish maw soaked in different seasoning was significantly different.Among them, the mixed seasoning group had the best adsorption effect on the flavor of fish maw, indicating that onion, ginger, and Sichuan pepper in the mixed seasoning group had synergistic effect on the overall odor of fish maw, which could not only cover up the unpleasant odor, but also enhance the overall flavor of fish maw.The second highest score was in the Sichuan pepper group, which was rich in terpenes, flavonoids, and amides, and could reduce the unpleasant odor in fish maw through sensory coverage (Chen et al., 2021;Yang et al., 2021).Terpenoids, shogaol, and gingerols were found to be closely related to the pungent and aromatic flavor of ginger (Krüger, Bergin, & Morlock, 2018).Onion's unique flavor characteristics are mainly due to its widespread sulfur compounds (Wang, Luca, & Edelenbos, 2019).The sulfide in onion and terpene in ginger also had a masking effect on the flavor of fish maw.The results showed that the unique flavor compounds of the seasoning had significant effects on the overall flavor of fish maw samples.The ginger group also had a better adsorption effect on fish maw flavor, mainly because ginger contained a variety of olefin compounds.Among them, β-sesquiterpene was a characteristic flavor compound in ginger (Peng et al., 2022), which had a strong volatile aroma and spicy smell, and had a certain stimulating effect on the senses, so it had adsorption effect on fish maw flavor.The adsorption effect of the deionized water group on the flavor of fish maw was small, and the effect of the unseasoned group was the worst.The main reason was that volatile compounds in the fish maw partially dissolve into the deionized water, which changed the flavor of the fish maw.The unseasoned group did not play the role of sensory masking on fish maw, so it had no significant effect on its flavor.

Statistical sorting method by R-index method
R-index is a method used to calculate the area of the subjects' perception curve, and the discrimination probability between paired samples represents the degree of difference between them (Feng et al., 2012).As shown in Table 3, R-index analysis found that the probability of selecting the unseasoned group instead of the Sichuan pepper group and the mixed seasoning group was 100%, indicating that the Sichuan pepper group and the mixed seasoning group had the best adsorption effect on the flavor of fish maw.According to R-value, there was no significant difference between unseasoned group and deionized water group, and no significant difference between Sichuan pepper group and mixed seasoning group.The results obtained by R-index analysis were consistent with those obtained by the rank sum test.

Analysis of E-nose
E-nose can be effective in obtaining information about volatile compounds in samples (Tiggemann et al., 2017).The E-nose was used to analyze the overall odor of six groups of fish gelatin soaked in different seasonings.Fig. 1a showed the PCA of E-nose.In Fig. 2a, PC1 and PC2 accounted for 64.2% and 15.93% of the total variation, respectively.This finding suggested that the first two PCs could be used in place of the original data.PCA could clearly distinguish the six samples, which were divided into two groups.Samples A, B, C, and D were located on the left side of the score plot, and samples E and F were located on the left side.It was demonstrated that the overall odor of samples A, B, C, and D are similar, but they were different from samples E and F. This might be due to different seasonings used in the soaking process.
The radar map of E-nose was shown in Fig. 1b, and there are some differences in the sensor response values of the six fish maws.The response values of R7 (inorganic sulfides), R9 (organic sulfides and aromatic compounds), and R6 (methyl groups) were more prominent, and the R7 sensor had the highest response value for the fish maw sample, indicating that the R7 sensor could better distinguish the volatile odors of the six groups.In addition, the response values of R7 and R9 sensors in Sichuan pepper group and mixed seasoning group were higher, indicating that the content of inorganic and organic sulfides in these two groups was higher than that in other groups.The response values of the other seven sensors were low, and the odor distribution of fish maw detected by R1 (aromatic components), R3 (ammonia and aromatic components), R4 (hydrides), R5 (short chain alkanes), and R10 (long chain alkanes) was almost completely coincident, indicating that the corresponding volatile odor substances were similar.Therefore, the major volatile compounds of the six groups of fish maw were inorganic sulfides, organic sulfides, aromatic compounds, and methyl groups.The results of E-nose were basically consistent with the results of sensory evaluation, but the E-nose was more accurate to distinguish different samples.1.
X. Miao et al.

HS-GC-IMS spectrum analysis of fish maw
Volatile components of fish maw in six groups (unseasoned group, deionized water group, onion group, ginger group, Sichuan pepper group and mixed seasoning group) were analyzed using FlavourSpec® flavor analyzer.Fig. 2a showed the 3D-topographic plot of HS-GC-IMS spectra of volatile organic compounds in fish maw of different seasoning groups.X-axis represented the retention time, y-axis represented the drift time, and z-axis represented the ion peak.Each point in the spectrum represented a volatile organic compound.The shade of color indicated the level of concentration.White meant low concentration, while red meant high concentration.Compounds might produce 1, 2 or more spots (representing monomers, dimers, or trimers) depending on the concentration and properties of the volatile components (Zhang, et al., 2023;Li et al., 2019).As shown in Fig. 2a, the 3D-topographic plot of fish maw in the six groups obtained by LAV analysis software were similar and difficult to distinguish volatile compounds with naked eyes.
As shown in Fig. 2b, the 3D-topographic plot of HS-GC-IMS was projected onto the 2D plane vertical view to obtain the 2D-topographic plot of volatile substances of fish maw in six groups.In order to reflect the differences of volatile components more intuitively between different groups, comparison spectra (Fig. 2c) between samples were drawn with the unseasoned group as the control.As shown in Fig. 2c, the characteristic spectral information of HS-GC-IMS of fish maw soaked in different seasonings was different.This might be related to differences in the characteristic volatile components of different seasonings.And as shown in the red box, some volatile components of fish maw soaked in different seasonings showed significant differences.In particular, the Sichuan pepper and mixed seasoning groups had higher levels of certain volatile compounds than the other four groups, and there was little difference in the number of volatile substances between the two groups.This was consistent with the PCA results of E-nose.

Qualitative analysis of HS-GC-IMS spectra of fish maw
The retention and migration times of characteristic volatile components were compared to calculate the retention index of each volatile component.In addition, n-ketone C4-C9 was used as the external Fig. 2. 3D-topographic plots (a), 2D-topographic plots (b: vertical view; c: difference view), qualitative analysis (d), gallery plot (fingerprint, e), PCA (f) and analysis of "nearest neighbor" fingerprints (g) of characteristic volatile organic.A, B, C, D, E, and F are the same as Table 1.standard for qualitative analysis of volatile components by matching GC-IMS database (Han et al., 2021;Song et al., 2021).The marked points shown in Fig. 2d represented the volatile components of qualitative analysis.A total of 95 volatile components (monomers and dimers of some substances) were identified using the software database, including 25 aldehydes, 23 olefins, 19 alcohols, 11 esters, 9 ketones, 3 acids, 2 sulfides, 1 furan, 1 ether and 1 ketoxime.The Monomer and dimer of the same substance had different forms, but had the same CAS number and chemical formula.The numbers in Table 4 corresponded to those in Fig. 2d.

Comparative analysis of HS-GC-IMS fingerprints of fish maw
In order to better compare the differences of volatile components of fish maw soaked in different seasonings.Three parallel tests were conducted on fish maw soaked in different seasonings, and all the peaks of the signals to be identified in the 2D-topographic of HS-GC-IMS were obtained.Finally, the fingerprints of volatile components of fish maw soaked in different seasonings were obtained (Fig. 2e).The rows in Fig. 2e represented samples, and the columns represented the same volatile component in different samples (the darker the red signal, the higher the content).As shown in Fig. 2e, the volatile components of fish maw soaked in different seasonings showed great differences (the red box in Fig. 2e).In the red rectangle area, the contents of cis-4-heptenal, trans-2-hexenal, 1-pentene-3-ol, trans-2-pentenal, 3-methylbutanol, 3occtanone, acetic acid, propionic acid, nonylaldehyde, ethyl propionate, and propional in the fish maw of the unsoaked group were higher than those in the other five groups.Most of the volatile compounds were fruity, oily and spicy, but there were also some pungent odors such as acetic acid, propionic acid and propionaldehyde, which could produce unpleasant odors.The contents of ethyl valerate, butyl acetate, butyl butyrate, styrene, butyl propionate and ethyl acetate in the fish maw of the deionized water group were higher than those in the other groups, and these volatile compounds mainly exhibited fruity and alcoholic aroma.In contrast, the onion group had higher contents of 1-pentanol and 1-penten-3-one, which were mainly introduced by onion (Wang, Luca, & Edelenbos, 2019).The contents of acetaldehyde and n-butanol were higher in ginger group and prickly ash group, respectively.However, the contents of camphene, dipropyl disulfide, D-camphor, 3-hydroxy-2-butanone, bornyl acetate, β-ocimene, α-terpinene, γ-terpinene, α-terpinolene, limonene, α-pinene; 6-methyl-5-hepten-2-one, 3-ene, 1,8cineole, 1-hexanol, and 2-n-pentylfuran in the fish maw of the mixed seasoning group were higher than those in the other groups (Li, Dadmohammadi, & Abbaspourrad 2022;Xi, Zhan, Tian, & Wang, 2019).The mixing of seasonings stimulates the formation of a variety of volatile compounds, which have a rich variety of odors.The mixed seasoning group effectively masked some of the unpleasant odors of the fish maw itself, which was consistent with the E-nose results.

Principal component analysis of volatile compounds
In this study, PCA of 95 volatile compounds from fish maw soaked in different seasonings was performed.As shown in Fig. 2f, the contribution of PC1 and PC2 were 57% and 19%, respectively.The cumulative contribution of the two principal components reached 76%.In general, the PCA model could be used as a separation model when the cumulative contribution reaches 60% (Wang et al., 2021).Therefore, PCA could better distinguish fish maw soaked in different seasonings, and there was no overlapping area among the six groups.This indicated that the volatile compounds of fish maw soaked in different seasonings had great differences.It could be seen from the aggregation degree of the groups, the difference between B, C and D was small, indicating that the volatile substances of the deionized water group, onion group and ginger group had little change.However, there were significant differences among E, F and A, indicating that the volatile compounds of Sichuan pepper group and mixed seasoning group varied greatly.This might be because the various volatile aromatic compounds in Sichuan pepper could change the flavor of fish maw.In addition, the "nearest neighbor" fingerprint analysis of fish maw soaked in different seasonings also showed the same results as PCA (Fig. 2g).The order of volatile compounds of fish maw soaked in different seasonings was F > E > D > C > B > A, indicating that volatile compounds in the Sichuan pepper group and mixed seasoning group had greater changes, while those in the deionized water group, onion group and ginger group had smaller changes.

Conclusion
In this study, sensory evaluation, E-nose, and HS-GC-IMS techniques were used to analyze the flavor changes of fish maw soaked in different seasonings.The volatile compounds of fish maw soaked in different seasonings were mainly organic sulfide and aromatic compounds.The Sichuan pepper group and the mixed seasoning group had better adsorption effect on the flavor of fish maw.The volatile components of fish maw soaked in different seasonings showed great differences.95 volatile organic compounds were identified by HS-GC-IMS technique.The contents of camphene, dipropyl disulfide, D-camphor, 3-hydroxy-2butanone, bornyl acetate, β-ocimene, α-terpinene, γ-terpinene, α-terpinolene, limonene, α-pinene, 6-methyl-5-hepten-2-one, 3-ene, 1,8cineole, 1-hexanol, and 2-n-pentylfuran in the fish maw of the mixed seasoning group were higher than those in the other groups.The mixing of seasonings stimulated the formation of a variety of volatile compounds, which had a rich variety of odors and were the major contributors to the flavor of fish maw.This study visualized the volatile flavor characteristics of fish maw soaked with different seasonings, and provided reference for producing high-quality fish maw and improving its flavor quality.

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.

Fig. 1 .
Fig. 1.PCA (a) and radar map (b) of E-nose.A, B, C, D, E, and F are the same as Table1.

Table 3 R
-index analysis of flavor adsorption effect of fish maw.

Table 4
Identification of volatile components.