Effects of Rewilding Aquaculture Time on Nutritional Quality and Flavor Characteristics of Grass Carp (Ctenopharyngodon idellus)

Abstract

Wild fish are preferred by consumers primarily for their superior sensorial qualities, including taste and texture. However, their limited availability often results in higher prices. Considering this, we explored the possibility of enhancing the quality of earthen pond aquaculture fish by transferring them to a near wild environment. This study investigated how rewilding time affects the physical properties, nutritional composition, and volatile profile of grass carp muscle. The results showed that compared to the 0M group, the crude protein content in grass carp muscle did not change significantly (p > 0.05) as the rewilding time increased to 6 months. Meanwhile, the significant increase in hardness and springiness (p < 0.05) indicated that the textural characteristics of muscle, which were key sensory and physical indices of muscle quality, were improved. Although the 6M group showed a 58.93% reduction in crude fat content compared to the 0M group, it retained the highest docosahexaenoic acid (DHA) content. Sensory evaluation demonstrated that as the rewilding time increased, the fishy and grassy odors of the rewilding grass carp diminished. Furthermore, cluster heatmaps and partial least squares discriminant analysis (PLS-DA) revealed that cultured grass carp and rewilding grass carp at three time points exhibited differences in various indicators. The variable importance in projection (VIP) showed that volatile flavor compounds (acetone, propionaldehyde-D, 1-penten-3-ol) and hardness were key factors in distinguishing between them. Therefore, extending the rewilding time can potentially enhance the acceptability of cultured grass carp by improving the physical properties, nutritional quality, and volatile profile of the muscle. This approach may provide a new pathway for fish aquaculture.

1. Introduction

To address global malnutrition and dietary imbalances, fish serve as a vital source of protein, vitamins, essential minerals, and unsaturated fatty acids, offering humans a diverse and high-quality selection. Grass carp (Ctenopharyngodon idellus) stand out as a primary commercial freshwater aquaculture species in Asia due to its high yield, rapid growth rate, and cost-effectiveness [1]. It is also favored by consumers for its nutritional richness and delicious meat. However, the rising demand has put additional strain on grass carp aquaculture, which primarily relies on earthen pond aquaculture methods. While earthen pond aquaculture has boosted freshwater fish production [2,3], it has also brought about negative consequences such as environmental degradation, water resource depletion, disease outbreaks, and concerns regarding product quality [4,5]. Changes in sensory quality have greatly affected consumers’ product choices. Among them, the flavor of aquatic products is mainly related to the content and type of amino acids and fatty acids [6]. Flavor substances are influenced by various sources including water conditions, fish feed, and environmental pollutants [7]. Therefore, it is necessary to further explore the flavor changes in grass carp under different aquaculture modes.

Market research has demonstrated the widely held belief that wild fish are healthier, more nutritious, and better tasting, whereas cultured fish are of poor quality [8,9]. For example, the high breeding density and single feed used in earthen pond aquaculture can degrade the aquaculture environment, potentially leading to slower fish growth and reduced meat quality. Recent studies indicate that fish cultured in flowing water systems can enhance muscle quality and nutritional value in an effective manner [10]. Fish cultured in such systems exhibit higher body size and higher survival rates compared to those in cultured earthen ponds [11]. Furthermore, they possess elevated protein content, reduced fat content, decreased muscle fiber diameter, and enhanced muscle hardness and springiness [12,13]. Whereas several studies compared the nutritional quality differences among different aquaculture systems [6,14]; in this study, we explored an innovative approach by re-locating earth pond grown fish in a near wild environment, a practice that can be named as rewilding aquaculture. Specifically, these fish are transferred to a controlled stream environment designed to simulate natural conditions. This approach allows the fish to grow in an environment that mimics their wild habitats, thereby integrating the principles of rewilding into aquaculture practices. Meanwhile, there are many differences between different aquaculture systems, and there are relatively few studies on the effects of rewilding aquaculture on the nutritional quality and flavor characteristics of grass carp.

Therefore, this study explored the differences in basic nutritional composition, amino acid and fatty acid content, and flavor between grass carp grown in earthen ponds and rewilded grass carp for 0, 2, 4, and 6 months. Furthermore, the key quality indicators of grass carp under different rewilding aquaculture times were selected by cluster analysis and PLS-DA. It aims to provide data for consumers’ decision and choice and provide a theoretical basis for the improvement, processing, and utilization of grass carp nutritional quality in the future.

2. Materials and Methods

2.1. Materials

Grass carp were sourced from the local market in Quzhou City (118°01′−119°20′ E, 28°14′−29°30′ N), Zhejiang Province. Grass carp, aged 16–18 months and measuring average weights and lengths of 1442 ± 550 g and 39 ± 6 cm, respectively, were initially maintained in earthen ponds with the density of 4 fish/m³. The consistent basic water quality conditions of earthen ponds are pH 7.0–7.5, O₂ content of 5.8–6.9 mg/L, and ammoniac nitrogen content of 0.2–1.1 mg/L. This initial group was designated as the 0M group.

Rewilding Conditions: The grass carp were subsequently transferred to a rewilding environment in simulated wild streams at the Hua Ruiyuan Family Farm in Quzhou. Rewilding periods were established at intervals of 2, 4, and 6 months, forming the groups labeled as 2M, 4M, and 6M, respectively. During rewilding, grass carp had unrestricted access to aquatic plants (such as algae and duckweed) supplemented with a controlled quantity (about 500 g/3 days) of commercial feed from Zhejiang Aohua Feedstuff Co., Ltd. (Zhejiang, China). The feed’s nutritional content included crude protein (≥30.0%), crude fat (≥3.0%), crude fiber (≤10.0%), coarse ash (≤15.0%), phosphorus (≥0.6%), lysine (≥1.45%), and moisture (≤12.0%). The initial stocking density was approximately 3 fish/m³ in the rewilding environment. The consistent basic water quality conditions are pH 6.2–7.1, O₂ content of 5.6–8.5 mg/L, and ammoniac nitrogen content of 0–0.11 mg/L.

Sampling Information: For the study, six grass carp were randomly selected from each group (0M, 2M, 4M, 6M), as shown in Figure 1. The selected fish were then transported alive to the Key Laboratory of Health Risk Factors for Seafood in Zhejiang Province for analysis.

2.2. Sample Preparation

The live grass carp was humanely euthanized through instantaneous puncturing, followed by the removal of the head, tail, and viscera. Subsequently, remaining parts of the fish underwent thorough washing with distilled water and were then stored in sealed bags at −40 °C for the subsequent determination of various indices.

2.3. Sensory Analysis

The sensory evaluation of grass carp fillets followed the method described by An et al. [15] with appropriate modifications. Fresh abdominal fillets were taken from four groups of grass carp after fishes were humanely euthanized at room temperature. Then, they were arranged on white paper, labeled as A, B, C, and D, and the remaining parts were refrigerated. The sensory evaluation panel consisted of 12 professionally trained members, including 6 males and 6 females, with ages ranging from 20 to 31 years. The sensory evaluation procedure followed ISO 10399:2017 [16] within one hour of euthanasia. The seven attributes that most accurately reflected the odor characteristics of grass carp were “fresh fish”, “fishy”, “grassy”, “earthy”, “alcohol”, “sulfury”, and “fatty”. Odor intensities were scored on the following scale: 0, no odor; 1, very weak; 2, weak; 3, moderate; 4, strong; and 5, very strong. Twelve trained professionals independently evaluated the fillets and recorded their results on the sensory evaluation form. The collected sensory evaluation forms were used to create radar charts representing the sensory differences among the four groups.

2.4. Determination of Nutritional Components

The grass carp fillets (n = 3) utilized for assessing basic nutritional indices were mainly sourced from the abdominal meat of grass carp. Moisture content of samples was determined using the 105 °C direct-drying method GB 5009.4-2016 [17] Ash content was determined using the muffle furnace volatilization constant weight method GB 5009.4-2016 [17]. Crude fat content was determined using the Soxhlet extraction method GB 5009.6-2016 [18], while crude protein content was determined using the micro-Kjeldahl method GB 5009.5-2016 [19].

2.5. Texture Analysis

Following the method outlined by Cheng et al. [20] with appropriate modifications, grass carp muscle tissue was cut into pieces (n = 9) measuring 10 mm × 10 mm × 5 mm. Texture analysis was conducted using the iTexture texture analyzer (Zhejiang Keqi Instrument Equipment Co., Ltd., Hangzhou, China). The parameters of the texture instrument were set as a P50 probe; pre-test speed = 5 mm/s; post-test speed = 5 mm/s; test speed = 1 mm/s; trigger force = 5 g; deformation rate = 50.00%; and dwell time = 5 s.

2.6. Determination of Fatty Acids

The fatty acid composition was determined through gas chromatography (GC-MS 7890B, US Agilent, Santa Clara, CA, USA) with a 7890B gas chromatograph. The fatty acid content in each group was calculated three times (n = 3) using the area normalization method. A single fatty acid methyl ester standard solution and a mixed fatty acid methyl ester standard solution were injected into a gas chromatograph to characterize the chromatographic peaks. Then, nitrogen gas (99.99%) conveyed the samples into a Supelco SP-2560 capillary column (100 m × 0.25 mm, film thickness 0.2 μm) with a 1.0 µL sample volume. The injection port and detector temperatures were 270 and 280 °C, respectively. The column temperature was held at 100 °C for 13 min, then raised to 180 °C at 10 °C min⁻¹ for 6 min, raised to 200 °C at 10 °C min⁻¹ for 20 min, raised to 230 °C at 4 °C min⁻¹, and held at this temperature for 10.5 min.

2.7. Determination of Free Amino Acids

Free amino acids (FAAs) were evaluated for three parallel samples (n = 3) of each group in accordance with the national food safety standard for the determination of amino acids in food. Trichloroacetic acid served as the extraction solvent for extracting FAAs, following the procedures outlined by Yu et al. [21]. Specifically, 5 g of muscle was precisely weighed and added to a 10 mL solution of 5% trichloroacetic acid. The resulting mixture was homogenized for 2 min, then subjected to centrifugation at 10,000× g and 4 °C for 10 min to obtain the supernatant. This process was repeated once; after which, all supernatants were diluted to a total volume of 25 mL. The determination of FAA content was found using the automatic amino acid analyzer (Agilent 1100, US Agilent, Palo Alto, CA, USA). The instrument used a 4.0 mm × 125 mm C18 column, the column temperature was 40 °C, and the buffer flow rate was 1.0 mL/min. The mobile phase A was 20 mmol/L sodium acetate, and the mobile phase B was prepared as V_{(20 mmol/L sodium acetate )}:V_(methanol):V_{(hexanitrile)} = 1:2:2.

2.8. Volatile Component Analysis by GC-IMS

Volatile component analysis of grass carp samples was performed using an integrated system comprising an Agilent 490 gas chromatograph (Agilent Technologies, Palo Alto, CA, USA) and IMS instrument (FlavourSpec^®, Gesellschaft für Analytische Sensorsystem mbH, Dortmund, Germany), coupled with an autosampler unit (CTC Analytics AG, Zwingen, Switzerland) for direct headspace sampling via a 1 mL air-tight heated syringe.

For processing, three fish fillets were selected from different groups; then, the fillets were chopped into minced meat state, and 2 g of the resulting mixture was transferred to a 20 mL headspace vial sealed with a magnetic cap. These samples were then incubated for 30 min at 40 °C with a rotating speed of 500 rpm. Following incubation, a heated syringe (85 °C) automatically injected a constant headspace (0.5 mL) into the injector. In the IMS unit, nitrogen gas (99.99%) conveyed the samples into an MXT-WAX capillary column (30 × 0.53 mm, film thickness 1 µm, Shimadzu Scientific Instruments, Kyoto, Japan) with a programmed flow rate of 2 mL/min for the initial 2 min, maintained at 2.00 mL/min for 8 min, then increased to 10 mL/min for 15 min, and finally to 100.00 mL/min for the remaining 5 min, resulting in a total runtime of 30 min. The ions of ionized analytes were directed to the drift tube at a constant temperature of 45 °C. The reported results represent the averages of three replicates. Substance identification relied on two-dimensional analysis of the retention index in gas chromatography and the relative migration time in the migration spectrum. Quantification involved calculating the relative content of the substance in different samples based on peak volume.

2.9. Statistical Analyses

All experiments were carried out in triplicates. Data were expressed as the means ± standard deviations. Statistical differences were analyzed using a one-way analysis of variance (ANOVA) followed by Duncan’s multiple range test. The hierarchical clustering analysis and partial least squares discriminant analysis (PLS-DA) were performed using MetaboAnalyst 6.0 online software (https://www.metaboanalyst.ca/, accessed on 15 February 2024). Other statistical analysis was performed using SPSS Statistics software (version 22.0, IBM, Chicago, IL, USA). A value of p < 0.05 was considered statistically significant.

3. Results and Discussion

3.1. Differences in Nutritional Indices and Texture Characteristics of Grass Carp with Different Rewilding Aquaculture Times

The nutritional and muscular characteristics of fish are influenced by external factors (such as farming temperature and water quality) and internal factors (such as genetics and metabolism) [9,22,23]. Thus, different aquaculture modes have a certain impact on the nutritional properties and muscle characteristics of cultured organisms. Table 1 presents the proximate composition and the texture parameters, hardness and springiness, of grass carp muscle under different rewilding aquaculture times. The results indicated that with an increase in rewilding aquaculture time, the moisture content in grass carp muscles gradually increased, with the 4M and 6M groups significantly higher than the 0M and 2M groups (p < 0.05). Meanwhile, the ash and crude protein content of grass carp throughout the rewilding aquaculture process showed no significant differences (p > 0.05). Interestingly, with the extension of rewilding aquaculture time, the crude fat content in the muscles of the 4M and 6M groups drastically decreased by 20.15% and 58.93%, respectively, compared to the 0M group. This may be associated with grass carp primarily consuming aquatic plants when adapting to a wild environment, as opposed to being fed nutritious artificial feed in earthen pond aquaculture mode [24]. At the same time, in order to adapt to the wild stream of rewilding aquaculture, grass carp need to consume more energy to cope with the increase in exercise, which may lead to less muscle fat [25]. Furthermore, Chen et al. [26] pointed out that within a certain range, the total fat content was not only related to the tenderness of fish muscle but also enhanced its flavor with an increase in fat content.

Muscle texture is an important descriptor of fish quality, also positively correlated with fish taste perception [27]. Hardness refers to the force required for food deformation, that is, the internal binding force to maintain the shape of food [28]. Springiness can reflect the combination of muscle tissue, which can reflect the taste of muscle fat [29]. Prior research demonstrated that consumers tend to favor wild fish for the excellent quality and firmer texture [30,31]. As shown in Table 1, earthen pond-cultured grass carp, after 4 months of rewilding aquaculture in streams, exhibited a significant (p < 0.05) increase in hardness and springiness, indicating that the textural characteristics of muscle were improved. Numerous studies also indicated that fish cultured under wild conditions tend to exhibit higher values of springiness and hardness compared to those cultured in earthen pond aquaculture [9,32]. Zhang et al.’s [33] study suggested that increasing the intensity of fish movement can enhance the binding force between muscle cells and reduce the diameter of muscle fibers, thereby improving muscle hardness and springiness.

The overall results indicated that rewilding aquaculture of grass carp in stream environments can lead to higher moisture content, lower fat content, and firmer and more elastic flesh. We speculated that the reason may be related to the wild stream culture model. Grass carp in the stream had a large range of activities and a large amount of movement, resulting in firmer meat than farmed grass carp. Thus, rewilding aquaculture may have a certain impact on the nutritional properties and muscle characteristics of fish.

3.2. Fatty Acid Composition and Content Analysis

Lipids in food not only serve as an energy source but also constitute a crucial supply of high-quality fatty acids for humans [34]. The relatively high levels of fish polyunsaturated fatty acids (PUFAs) have drawn consumer attention, especially omega-3 long-chain PUFAs. Thirty fatty acids, including twelve saturated fatty acids (SFAs), eight monounsaturated fatty acids (MUFAs), and ten PUFAs, were detected in the muscles of grass carp under different rewilding times (Table 2). Considering the crude fat content in grass carp muscle, the actual content of saturated fatty acids in the 0M, 2M, 4M, and 6M groups were 1.715, 1.621, 0.848, and 0.715 g/100 g, respectively. Moreover, with increasing rewilding time, the content of SFAs, MUFAs, and PUFAs in grass carp exhibited a decreasing trend. The most important for human’s healthy omega-3 long-chain PUFAs include eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA). EPA and DHA content are often considered a key indicator for evaluating the nutritional value of fatty acids because of their various physiological functions that are beneficial to the human body, such as the maintenance of cardiovascular health, anti-inflammatory effects, and regulation of the immune system [35]. It was noteworthy that the DHA (C22:6, cis-4,7,10,13,16,19) content in the muscle of the 6M group was significantly (p < 0.05) higher than the other groups, reaching 1.440 g/ 100 g in the oil extract and 0.043 g/ 100 g in the muscle. However, the EPA content was lower than that in the 0M and 2M groups. Additionally, some studies suggested that the EPA and DHA content in cultured fish was similar to or higher than that in wild fish [32,36]. The results of this study suggest that with the extension of rewilding time, the fatty acid composition in grass carp showed minimal differences, where EPA and DHA decreased first and then increased. However, influenced by the decrease in crude fat content, there was an overall decreasing trend in fatty acid content. Furthermore, fatty acids contribute significantly to the flavor of fish [6], and thus, further exploration is needed to understand the impact of rewilding aquaculture time on the flavor of grass carp.

3.3. Free Amino Acid Composition and Content Analysis

Amino acid is an important indicator of fish meat nutrition value [37]. The composition and content of amino acids in muscles are typically influenced by aquaculture methods, feed or prey, and the aquaculture time [32]. As shown in Table 3, a total of 16 common amino acids were identified in four fish groups, including seven essential amino acids (EAAs), four delicious amino acids (DAAs), and five other amino acids (Table 3). The ratios of EAAs to total amino acids in grass carp samples were 0.77, 0.41, 0.41, and 0.41, respectively. The ratios of essential amino acids to non-essential amino acids in grass carp of 0M, 2M, 4M, and 6M groups were 1.43, 0.75, 0.74, and 0.74, respectively, all exceeding the ideal model standards proposed by FAO/WHO (ΣEAA/ΣTAA approximately 0.4, ΣEAA/ΣNEAA greater than 0.6) [38]. Therefore, the amino acid composition in the muscle of grass carp in all four groups met human nutritional requirements. It directly affects taste and indirectly participates in the development of flavor [39]. Studies suggested a positive correlation between animal protein flavor and DAAs (aspartic acid, glutamic acid, glycine, and proline) [40,41]. On the other hand, histidine, lysine, valine, methionine, phenylalanine, leucine, and tyrosine are considered unpleasant bitter-tasting amino acids [42]. This study found an overall decrease in the content of umami and bitter-tasting amino acids with increasing rewilding aquaculture time. Wang et al. [43] discovered higher umami levels in wild Chinese mitten crab compared to the cultured group. Other research indicated that both earthen ponds and recirculating aquaculture systems can enhance umami and increase DAA content compared to wild Yellow River carp [32]. However, the results are inconsistent with the expected outcomes, possibly due to the uncertainty in grass carp food sources under rewilding aquaculture mode.

3.4. Sensory Analysis

Based on the quantitative descriptive analysis results from the sensory panel, a radar chart depicting the sensory evaluation of grass carp under different rewilding aquaculture times was generated (Figure 2a). The overall odor of grass carp was relatively mild, with minor differences in some flavor attributes (including fresh fish, sulfury, and fatty flavors) among the groups, with average intensities for each attribute below 3.5. The intensities of sulfury and fatty flavors were even below 1.5, almost negligible in the overall flavor profile of grass carp. However, significant differences were observed in fishy, grassy, earthy, and alcoholic flavors. From the sensory scores, the earthen pond-cultured grass carp (0M) has scores for grassy and earthy flavor 30% and 48% higher than those of the grass carp rewilded for 6 months, indicating that appropriate rewilding aquaculture helped reduce the fishy and earthy flavors in cultured grass carp. The formation of off-flavors in cultured freshwater fish can be attributed to aquaculture and post-harvest aspects. During the cultured stage, the generation of off-flavors in cultured freshwater fish is mainly caused by factors such as water quality (including pollutants like 2,4-heptanedial and hexanal) and feed ingredients [7]. Wang et al. [44] pointed out that water quality conditions played a crucial role because pollutants in aquatic ecosystems, such as 2,4-heptanedial and hexanal, could lead to the development of off-flavors in cultured freshwater fish. Additionally, the lipid content in feed ingredients was also associated with the intensity of fishy flavors in fish meat [45]. Therefore, rewilding aquaculture reduced the possibility of grass carp accumulating pollutants and consuming high-fat feed, thereby improving the flavor of fish. At the same time, compared with the real wild environment, the water quality, temperature, pollutants, and food sources involved in grass carp rewilding aquaculture are more controllable.

3.5. GC-IMS Analysis

According to the retention time, migration time, and peak intensity of different volatile components, a top view of the GC-IMS 3D topography of grass carp volatile components was obtained by normalizing the ion migration time and reactive ion peak (RIP) position. (Figure 2b). To compare the differences most obvious, the spectrum of 0M group was selected as reference and the spectra of other samples were deducted from the reference. If the concentration of two groups was consistent, the background after deduction was white. Red dots indicated that the concentration of the substance was higher than the reference, while blue indicated that the concentration of the substance was lower. Most of the signals were concentrated in the retention time range of 0–700 s and the drift time range of 6.0–10.0. The volatile flavor compounds of the 0M group were different from those of the other three groups, and most compounds showed a decreasing trend, which may be related to the difference in fatty acid production found in the above study. It showed that the volatile odor components of grass carp will be affected by the rewilding aquaculture times, which may be due to the change in lipid content [46].

To validate the qualitative results of GC-IMS, the relative contents of 35 volatile organic compounds (VOCs) (combining monomers and dimers) were identified from three batches of samples in each of the four grass carp groups, as shown in Table 4. These compounds were further categorized into thirteen alcohols, eleven aldehydes, six ketones, one acid, three esters, and one ether. Some single compounds may produce multiple signals or spots (dimers or trimers) due to different concentrations of the compounds [47]. It is noteworthy that alcohols, aldehydes, and ketones in the samples exhibit rich olfactory attributes. Flavor thresholds values (FTVs) are the lowest concentrations of compounds that human perception can perceive through the tongue or taste buds [48]. The lower the compound threshold, the more sensitive humans are to its flavor. Relatively low thresholds were found for 1-hexanol (threshold: 0.5 mg/L), hexanal (threshold: 4.5–5 ug/L), 1-penten-3-ol (threshold: 350–400 ug/L), nonanal (threshold: 0.34 µg/L), 1-octen-3-ol (threshold: 1.5 ng/g), and 2-butanone (threshold: 1.55–7.76 mg/L). These compounds are commonly considered to be off-flavors in freshwater fish, exhibiting fishy, grassy, earthy, alcoholic, and other odors. As can be seen in Table 4, there was an overall trend of decreasing signal intensities detected for the above compounds with increasing rewilding time, indicating a reduction in odorants. This is consistent with the results of sensory evaluation (Figure 2a). However, comparing the differences in odor characteristics among grass carp with different rewilding aquaculture times based solely on Table 4 was complex and not intuitive. Therefore, it was necessary to establish VOC fingerprint profiles based on the relative contents of VOCs in each group of grass carp.

As shown in Figure 2c, VOC fingerprints of grass carp with different rewilding aquaculture times were constructed based on 35 differential VOCs in the samples. The results indicated significant differences in odor composition between non-rewilded grass carp (0M) and those rewilded for 2, 4, and 6 months, with an overall decrease in odor content. Under the earthen pond aquaculture mode, the 0M group of grass carp contained higher levels of 1-octen-3-ol, 1-hexanol-M, 1-butanol-M, 1-propanol-M, and acetone, represented by a deep red fingerprint, which was consistent with the results of Xiao et al. [49], while these flavor components gradually decreased with increasing rewilding time. The concentration of 3-hydroxy-2-butanone, heptanal, butyraldehyde (D), and (E)-2-pentenal was higher in 2M and lower in other groups. The concentration of 1-pentanol (M and D) and 1-penten-3-ol was higher in the 0M or 2M group, reached the peak in 2M fish, and then gradually decreased. The concentration of hexanal (D) and pentanal was higher in the 2M and 4M groups. The concentration of acetone and 2-butanone (M and D) was higher in groups 0M–2M, and concentrations of propionaldehyde (D) and butyraldehyde (M) were the highest at 2M and the lowest at 6M.

Specifically, alcohols are often described as an important component of fish volatiles, as they are closely related to the characteristic fatty flavor of fish [47]. The main volatile alcohol was ethanol (M and D), which was similar to that of dry-cured fish. The threshold of alcohols was high [50], and although the relative content of ethanol was the highest, its contribution to flavor was not significant. The alcohol flavor of fish could be derived from the oxidative decomposition of lipids or the reduction synthesis of carbonyl groups [51]. It is worth noting that there is a significant decrease in the relative content of 1-octen-3-ol. It has a grassy aroma and is a common off-flavor in freshwater fish. Additionally, the relative content of 1-octen-3-ol decreases with increasing rewilding time, and its content can be used as one of the criteria for freshness determination [52]. Moreover, aldehydes, the main products of lipid oxidation, appear to be the main contributors to the flavor of fish as they show low olfaction thresholds and distinctive odor characteristics [47]. Saturated linear aldehydes such as valeraldehyde and hexanal have been detected in a variety of fish and have been proven to be important volatile flavor components [53,54]. Propionaldehyde and butyraldehyde have not been mentioned in most studies, and their sources may be caused by endogenous enzymes or microorganisms [55]. The aldehydes with the highest relative content in four groups of grass carp were propionaldehyde and hexanal. The relative content of most aldehydes reached the maximum at 2M and then gradually decreased. It showed that the volatile flavor of grass carp was the strongest at 2M. The increase in volatile flavor of 2M may be related to the oxidation of polyunsaturated fatty acids such as linoleic acid and linolenic acid.

3.6. Correlation Analysis and PLS-DA

The hierarchical clustering heatmaps (Figure 3a) were used to analyze the differences in nutritional quality and flavor characteristics of grass carp under different rewilding times. The clustering results indicated that the 0M and 2M groups exhibit similarities in nutritional quality and flavor characteristics, while the 4M and 6M groups gradually diverge from the 0M group. In the 0M and 2M groups, grass carp exhibited higher proportions of crude fat, grassy flavor, earthy flavor, 1-octen-3-ol, 2-heptanone, 2-pentanone, 1-penten-3-ol, and acetone. Aldehydes, alcohols, ketones, and other low-molecular-weight volatile compounds are formed through the extended oxidation of long-chain unsaturated fatty acids and lipid breakdown [50]. Their elevated concentrations could potentially influence the flavor of fish products [56]. Leucine, phenylalanine, isoleucine, 1-propanol-M, 1-propanol-D, proline, acetone, 1-penten-3-ol, 2-pentanone, 1-octen-3-ol, 2-heptanone, and similar compounds gradually decreased with increasing rewilding time. One of the characteristic aromas of fish at low levels is 1-octen-3-ol [57], but at a high level, it produces a fishy flavor, which is a characteristic compound causing off-flavors [58]. The reduction in the content of 1-octen-3-ol indicates that wild cultivation effectively reduces the presence of fishy odors in grass carp muscles.

To further distinguish the differences among grass carp with different rewilding aquaculture times, a PLS-DA was conducted (Figure 3b,c). Taking the rewilding time of grass carp as the independent variable and utilizing basic physicochemical indicators, fatty acids, amino acids, sensory scores, and volatile flavor compounds, as dependent variables, a partial PLS-DA model was constructed. PLS-DA is a supervised discriminant analysis method for processing classification and discriminant problems [59,60]. The farther away the two samples are from each other in the score table, the greater the difference between the two samples [61]. The contributions of component one and component two were divided into 33.1% and 51.1%, respectively, with a cumulative total of 84.2%, reflecting the overall variance in the entire dataset. Based on the scores plot graph derived from the PLS-DA model (Figure 3b), the closer the values of each indicator, the more similar the two sample groups. Similar to the results of hierarchical clustering analysis, the PLS-DA identified the 6M group as distinct from the other groups, forming two major clusters, indicating significant differences between the 6M group and the others. Furthermore, the 2M group still exhibited some similarity to the 0M group, with distinct separation occurring only after 4 months of rewilding.

The variable importance in projection (VIP) is commonly employed to assess the strength and explanatory power of the impact of changes in various indicators on the classification and differentiation of different sample groups. Generally, a VIP value exceeding 1.0 is utilized as the screening criterion. According to the results, acetone, propionaldehyde-D, 1-penten-3-ol, 1-propanol-M, 1-octen-3-ol, hexanal-M, ethanol-M, hardness, and 1-propanol-D play a primary role in identifying earthen pond aquaculture mode and rewilding aquaculture. Specifically, flavor substances contribute significantly to the discrimination process, with hardness providing consumers with an intuitive judgment. The remaining indicators can also be considered from the perspective of rapid detection technology development.

4. Conclusions

The results of this study suggest that in terms of nutrients, as the time of rewilding aquaculture increased, the total crude fat content decreased, while the content of high-quality fatty acids EPA and DHA decreased first and then increased. At the same time, the hardness and springiness of grass carp muscle increased significantly, aligning with the texture characteristics favored by consumers in fish. The amino acid composition met human nutritional requirements. As far as flavor was concerned, there was an overall decrease in the content of umami and bitter-tasting amino acids with increasing rewilding time. The effect of different rewilding times on the flavor of grass carp was examined using GC-IMS, and the results showed that the content of the main volatile flavor components related to off-flavors showed an overall decreasing trend. In summary, rewilding had an effect on the muscle physical properties, nutritional components, and volatile substances of grass carp. The results suggest significant changes were observed when the rewilding time reached 4 months. On whether the rewilding aquaculture and the selection of rewilding time can contribute to improve the quality and flavor of fish, this study provides a certain reference. However, the optimal rewilding time may vary depending on the fish species. Therefore, further exploration is needed to determine the rewilding time for other fish species.