Effects of Submerged Macrophytes on the Growth, Morphology, Nutritional Value, and Flavor of Cultured Largemouth Bass (Micropterus salmoides)

Aquaculture environment plays important roles in regulating the growth, morphology, nutrition, and flavor of aquatic products. The present study investigated growth, morphology, nutrition, and flavor formation in largemouth bass (Micropterus salmoides) cultured in the ponds with (EM group) and without (M group) the submerged macrophytes (Elodea nuttallii). Fish in the EM group showed a significantly greater body length, higher growth rate, and lower hepatosomatic index than those in the M group (p < 0.05). Moreover, compared with fish in the M group, those in the EM group showed improved muscle quality with significantly elevated levels of crude protein, total free and hydrolysable amino acids, and polyunsaturated fatty acids (p < 0.05). Specifically, certain amino acids related to flavor (Glu, Asp, Ala, and Arg) and valuable fatty acids (C18:2, C18:3n3, C20:3n3, and C22:6) were more abundant in the EM group (p < 0.05). In addition, the levels of 19 volatile (p < 0.05) were significantly higher in the EM group than in the M group. Therefore, E. nuttallii significantly improved growth, morphological traits, nutritional components, and characteristic flavor in largemouth bass, indicating the superior nutritional value and palatability of fish cultured with submerged macrophytes.


Introduction
China is one of the largest producers and consumers of aquatic products in the world, accounting for over 60% of the global aquaculture output [1]. Among aquatic products, the annual production of largemouth bass (Micropterus salmoides)-an economically important freshwater aquaculture species [2]-reached about 0.62 million tons [3] with an output value of over $1.76 billion in 2020. This species is native to lakes and rivers in North America [4]. Owing to its rapid growth, delicious flavor, and lack of intermuscular bones, largemouth bass has become widely popular in China since its introduction in 1983 [5]. At present, traditional pond systems remain the only acceptable mode of largemouth bass culture in China [6]. During pond culture, artificial compound feed is supplied, which leads to eutrophication, thereby promoting the outbreak of bloom-forming cyanophytes and diseases and ultimately deteriorating the quality of aquaculture products [7,8]. Therefore, these challenges must be addressed by developing and promoting new ecological aquaculture models.
Ecological aquaculture not only protects the environment but also fulfills the market demand. In recent times, with increased affluence, the focus of consumption has shifted Samples were collected every month to record growth performance. Body weight, body length, and liver and visceral weights of 30 fish in each group were measured after 24 h of starvation, and the fish anaesthetized with 100 mg·L −1 MS-222. At the end of the experimental period, dorsal muscles of 12 fish in each group were sampled, and stored at −80 • C for subsequent analysis. Water samples were collected from the experimental ponds to determine water quality parameters. All animal experiments conformed to the ARRIVE guidelines and were performed following the U.K. Animals (Scientific Procedures) Act, 1986, and the associated guidelines of the EU Directive 2010/63/EU for animal experimentation.

Water Quality Determination
Dissolved oxygen (DO) and pH were measured in situ using a portable multimeter (HQ30D; HACH, Ames, IA, USA) and the YSI Professional Plus system (YSI Inc., Yellow Springs, OH, USA). To test quality, triplicate water samples from each of the treatment sets were transferred to 500 mL polyethylene bottles and the physicochemical parameters of total nitrogen (TN), total phosphorus (TP), and chemical oxygen demand (COD Mn ) were analyzed as described previously [25]. All samples were filtered using Whatman filter papers with a pore size of 0.45 µm before laboratory analyses.

Biological, Color and Muscle Nutrients Measurements
In the present study, we calculated the weight gain rate (WGR), specific growth rate (SGR), and hepatosomatic index (HSI) as parameters reflecting growth performance, as follows.
The approximate composition of muscles was investigated following the national standard methods, with three parallel measurements per group. Moisture content was determined according to the AOAC Official Method 930.15 (drying at 105 • C to a constant weight). Ash content was measured according to the AOAC Official Method 942.05 (burning at 550 • C in a muffle furnace) [27]. Crude protein content was determined according to the AOAC Official Method 968.08 (Kjeldahl nitrogen determination method) [28], and crude fat was determined according to the AOAC Official Method 996.06 (Soxhlet extraction method).
For fatty acid analysis, fatty acid methyl esters (FAMEs) were prepared by transesterification with boron trifluoride and methanol, then dissolved in hexane, and the upper organic phase was collected for analysis with an Agilent 7820 A Gas Chromatograph (Agilent Technologies, Inc., Santa Clara, CA, USA) [29].
To determine amino acid content, muscle samples were freeze-dried and ground to powder. Next, 0.1 g samples were accurately weighed and used for amino acid determination. Briefly, the samples were treated with 6 M HCl for acid hydrolysis at 120 • C for 22 h, and then then neutralized with NaOH, and the supernatant was collected for analysis. Free amino acid were adjusted to an appropriate volume with 5% trichloroacetic acid, mixed well, then allowed to stand for 2 h and filtered. Finally, the supernatant was collected for analysis. Amino acid analyses were performed using high-performance liquid chromatography (HPLC) (Ag 1260 HPLC, American Agilent Company), according to the method described by Harimana [5].

Comparison of Fish Muscle Volatile Substances
Volatile compounds were identified using GC-IMS [30]. Muscle samples from each group were weighed and chopped evenly. Each sample was analyzed in triplicate to ensure the reliability of results. Briefly, 3 g samples in 20 mL headspace bottles were randomly selected. The analytical conditions were as follow: headspace incubation = 15 min, temperature = 60 • C, speed = 500 rpm, injection volume = 500 µL, and syringe temperature = 110 • C. GC conditions were as follows: chromatographic column = MXT-5 (15.00 m × 0.53 mm, 1.00 µm i.d.), column temperature = 60 • C, run time = 20 min, and carrier gas = N 2 (purity ≥ 99.999%). The initial flow rate of the carrier gas was 2 mL·min −1 for 2 min, which was increased to 100 mL·min −1 , and the total run time was 20 min. IMS conditions were as follows: temperature = 45 • C and drift gas flow rate = 150 mL·min −1 . The retention index (RI) of each compound was calculated. The analytical software supporting the measurement instruments were vocal, three plug-ins (Reporter, Gallery Plot, and Dynamic PCA), and GC-IMS Library Search, which can analyze samples from different perspectives. Spiked and non-spiked samples were measured five times in parallel to calculate the recovery rate and relative standard deviation (RSD).

Statistical Methods and Data Processing
Data collated using Microsoft Excel were expressed as mean ± standard deviation (SD). In SPSS v26.0. (IBM Corporation, Armonk, NY, USA), t-test was performed to determine significant differences between the groups. A p < 0.05 indicated significant (*), p < 0.01 indicated highly significant (**), and p < 0.001 indicated extremely highly significant (***) difference.

Growth Performance and Morphological Characteristics
The monthly growth performance of fish is summarized in Table 1. On day 30, there were no significant differences in BL, BT, BW, WGR, or SGR (p > 0.05), whereas fish in the EM group showed a significantly lower HSI than those in the M group (p < 0.05). On day 60, compared with fish in the M group, those in the EM group showed a significantly higher BW and SGR (p < 0.05) and a lower BT and HSI (p < 0.05). One month later (at 90 days), the BL, BW, WGR, and SGR of fish in the EM group significantly increased, while HSI continued to decrease significantly (p < 0.05). Moreover, fish in the EM group were slender and presented a green body ( Figure 1), with darker dorsal skin, as evidenced by significantly lower L* values ( Table 2, p < 0.01). Regardless of the origin (dorsal or abdominal skin), significant differences were observed in a* and b*.    14.50 ± 1.13 *** 5.97 ± 1.14 6.25 ± 0.71 *** 2.15 ± 0.47 Note: p < 0.01 indicated highly significant (**), and p < 0.001 indicated extremely highly significant (***) difference.

Water Quality and Dominant Phytoplankton
Water chemical indices and dominant phytoplankton, including Cyanophyta and Chlorophyta, are shown in Figure 2. At the end of the 90-day experimental period, significant differences in four water quality indices (TN, TP, DO, and CODMn) and dominant phytoplankton reflected the variations in ecological factors for aquaculture water between groups. Compared with values in the M group, TN, TP, and CODMn in the EM group were significantly decreased, while DO was significantly increased (p < 0.05). Furthermore, in  Note: p < 0.01 indicated highly significant (**), and p < 0.001 indicated extremely highly significant (***) difference.

Water Quality and Dominant Phytoplankton
Water chemical indices and dominant phytoplankton, including Cyanophyta and Chlorophyta, are shown in Figure 2. At the end of the 90-day experimental period, significant differences in four water quality indices (TN, TP, DO, and COD Mn ) and dominant phytoplankton reflected the variations in ecological factors for aquaculture water between groups. Compared with values in the M group, TN, TP, and COD Mn in the EM group were significantly decreased, while DO was significantly increased (p < 0.05). Furthermore, in the M group, cyanobacteria accounted for 60% of all phytoplankton, with a density of 1.36 × 10 7 cells·L −1 , which was significantly higher than that in the EM group (2.28 × 10 5 cells·L −1 ; p < 0.01). More specifically, nearly 60-fold difference was noted between the two groups.

Nutritional Components
The proximate compositions of samples varied (Table 3). All samples were rich sources of proteins. The crude protein content of samples in the M and EM groups was respectively 21.13% and 23.07% (p < 0.05). Moisture content was significantly higher in the M group (p < 0.01). Ash and crude fat content did not significantly differ between the two groups (p > 0.05). the M group, cyanobacteria accounted for 60% of all phytoplankton, with a density of 1.36 × 10 7 cells·L −1 , which was significantly higher than that in the EM group (2.28 × 10 5 cells·L −1 ; p < 0.01). More specifically, nearly 60-fold difference was noted between the two groups.

Nutritional Components
The proximate compositions of samples varied (Table 3). All samples were rich sources of proteins. The crude protein content of samples in the M and EM groups was respectively 21.13% and 23.07% (p < 0.05). Moisture content was significantly higher in the M group (p < 0.01). Ash and crude fat content did not significantly differ between the two groups (p > 0.05). Seventeen free and hydrolysable amino acids were detected in different samples (Table 4). Levels of free amino acids, including Gly, Thr, Tyr, Phe, and Ile, in the muscles of Seventeen free and hydrolysable amino acids were detected in different samples ( Table 4). Levels of free amino acids, including Gly, Thr, Tyr, Phe, and Ile, in the muscles of largemouth bass were significantly higher in the EM group than in the M group. Moreover, levels of essential, no−essential, and total free amino acids significantly differed between the two groups (EM > M). Among hydrolysable amino acids, Glu content was the highest in different samples, and Glu content in the EM group was significantly higher than that in the M group. In addition, Asp, Ala, Arg, Ser, His, Thr, Val, Phe, IIe, and Leu levels were higher in the EM group than in the M group.

Volatile Compounds
The entire spectrum representing total volatile substances was presented as twodimensional topographical visualization. Figure 3 shows significant differences in the gas-phase ion migration spectra of muscle samples. The concentration of volatile substances was significantly lower in the M group than in the EM group. We used a different comparison system to visualize the differences between samples. Taking M1 as the reference, the remaining spectral values were deducted from the signal peaks in M1 to obtain the differences in spectra (Figure 4). Substances with levels lower than those in M1 are shown in blue (region A), whereas those with levels higher than those in M1 are shown in red (region B). If the levels of volatile substances are comparable, the background after deduction is white. The deeper the color, the greater the difference. Differences between Figures 3 and 4 clearly demonstrate that the concentration of volatile organic compounds was consistently higher in the EM group. The galleryplot plug-in of the LAV software was used to automatically generate fingerprints of all peaks for determining characteristic differences in volatile substances. As shown in Figure 5, substances related to flavor presented characteristic and common peak areas in the two groups. Regions A and B in Figure 5 represent the characteristic peak areas of the M and EM group, respectively. Therefore, the flavor of samples significantly differed between the M and EM groups.   Next, principal component analysis (PCA) was applied to understand the correlations in largemouth bass muscle samples. PC1 explained 36% sample variance, whereas PC2 explained 28% sample variance ( Figure 6). Based on these data, the samples were divided into two groups, and the between-group difference was greater than the within-group difference. Therefore, GC-IMS is suitable to distinguish largemouth bass from different culture models.
ized reaction peak. The y-axis represents the retention time of gas chromatography, and the x-ax represents the ion relative drift time. Points on both sides of the reaction peak represent volatile organic compounds. Colors indicate the concentration of substance, with white and red represen ing a low and high concentration, respectively. The deeper the color, the higher the concentration For the interpretation of references to colors in this figure legend, please refer the web version of this article.   Next, principal component analysis (PCA) was applied to understand the correlations in largemouth bass muscle samples. PC1 explained 36% sample variance, whereas PC2 explained 28% sample variance ( Figure 6). Based on these data, the samples were divided into two groups, and the between-group difference was greater than the withingroup difference. Therefore, GC-IMS is suitable to distinguish largemouth bass from different culture models. In the present study, 55 volatile compounds were identified, of which 54 were qualitative substances, primarily comprising aldehydes, alcohols, ketones, acids, esters, and miscellaneous compounds (Table 6)  Next, principal component analysis (PCA) was applied to understand the co tions in largemouth bass muscle samples. PC1 explained 36% sample variance, w PC2 explained 28% sample variance ( Figure 6). Based on these data, the samples divided into two groups, and the between-group difference was greater than the w group difference. Therefore, GC-IMS is suitable to distinguish largemouth bass fro ferent culture models. In the present study, 55 volatile compounds were identified, of which 54 were tative substances, primarily comprising aldehydes, alcohols, ketones, acids, ester miscellaneous compounds (Table 6). Twenty-four aldehydes accounted for 43.64%   Each row in the figure represents all signal peaks selected in a sample, and each column represents signal peaks of the same volatile compounds in different samples. -M and -D indicate the monomer and dimer of the same substance, and these are presented behind some substance. Numbers indicate unidentified peaks. Note: p < 0.05 indicated significant (*), p < 0.01 indicated highly significant (**), and p < 0.001 indicated extremely highly significant (***) difference.

Correlation Analysis
To better understand the key factors affecting the flavor and nutritional quality of largemouth bass, we performed Spearman's correlation analysis. As shown in Figure 7A, the abundance of all volatile substances analyzed was positively correlated with that of DO and chlorophyta, but negatively correlated with that of TN, TP, and COD Mn . In addition, 2-ethyl-1hexanol-D and 2-ethyl-1-hexanol-M were significantly and positively correlated with DO and Chlorophyta but negatively correlated with TN and TP (p < 0.05). Cyanobacteria and COD Mn are important biotic and abiotic factors affecting aquatic animals, respectively, and they were significantly but negatively correlated with 68% of the analyzed volatile substances (nonanal-M, nonanal-D, octanal-D, oct-1-en-3-ol-M, 3-furanmethanol, benzaldehyde-M, heptanal-D, oct-1en-3-ol-D, 2-octanol, 2-methylbutanal-D, 3-methylbutanal-D, 2-Butanone, 3-hydroxybutan-2one-D,3-hydroxybutan-2-one-M, and (E)-3-penten-2-one-D). Regarding nutritional components ( Figure 7B), free and hydrolysable most amino acids were significantly and positively correlated with DO and Chlorophyta and significantly but negatively correlated with TN and TP (p < 0.05). ΣPUFAs (C18:2 and C18:3n3) were significantly and positively correlated with DO and Chlorophyta (p < 0.05). C20:3n3 levels were significantly but negatively correlated with cyanobacteria and COD Mn (p < 0.05), and C22:6 levels were significantly but negatively correlated with TN (p < 0.05). Based on these results, cyanobacteria, Chlorophyta, COD Mn , DO, TN, and TP may be the key biotic and abiotic factors affecting the flavor and nutritional value of largemouth bass.

Effects of Submerged Macrophytes on the Growth and Morphology of Largemouth Bass
Water quality is a critical factor in aquaculture, as poor-quality water can significantly impede growth and production [31]. In the present study, the higher content of TN, TP, and CODMn in the M group, which lacked submerged macrophytes, led to cyanobacterial outbreak (1.36 × 10 7 cells·L −1 ). Conversely, this phenomenon was not observed in the EM group, which comprised submerged macrophytes. TN and TP are the two most important indices of the eutrophication of water bodies [32], while CODMn is an indicator of organic pollution [33]. Higher values of these indices promote the proliferation of phytoplankton and outbreak of cyanobacterial blooms [34]. Cyanobacteria can produce abundant toxic secondary metabolites, such as dermatoxins, hepatotoxins, and cytotoxins [35], which affect the feeding, growth, and immunity of exposed cultured species [36]. In our experiment, fish in the EM group showed a significantly higher growth rate, corroborating previously reported experimental findings. For instance, Yao showed that the inclusion of live submerged macrophytes in tanks improved the growth of Macrobrachium nipponense [37]. Meanwhile, in the present experiment, largemouth bass cultured in the presence of submerged macrophytes were slender, with a green body. Our observations are consistent with reported findings in largemouth bass cultured in an aquaculture system using land-based containers with recycled water [6]. Moreover, fish in the M group showed a significantly higher HSI, suggesting that largemouth bass cultured in the con-

Effects of Submerged Macrophytes on the Growth and Morphology of Largemouth Bass
Water quality is a critical factor in aquaculture, as poor-quality water can significantly impede growth and production [31]. In the present study, the higher content of TN, TP, and COD Mn in the M group, which lacked submerged macrophytes, led to cyanobacterial outbreak (1.36 × 10 7 cells·L −1 ). Conversely, this phenomenon was not observed in the EM group, which comprised submerged macrophytes. TN and TP are the two most important indices of the eutrophication of water bodies [32], while COD Mn is an indicator of organic pollution [33]. Higher values of these indices promote the proliferation of phytoplankton and outbreak of cyanobacterial blooms [34]. Cyanobacteria can produce abundant toxic secondary metabolites, such as dermatoxins, hepatotoxins, and cytotoxins [35], which affect the feeding, growth, and immunity of exposed cultured species [36]. In our experiment, fish in the EM group showed a significantly higher growth rate, corroborating previously reported experimental findings. For instance, Yao showed that the inclusion of live submerged macrophytes in tanks improved the growth of Macrobrachium nipponense [37]. Meanwhile, in the present experiment, largemouth bass cultured in the presence of submerged macrophytes were slender, with a green body. Our observations are consistent with reported findings in largemouth bass cultured in an aquaculture system using land-based containers with recycled water [6]. Moreover, fish in the M group showed a significantly higher HSI, suggesting that largemouth bass cultured in the conventional model produced excess body energy, which led to lipid and glycogen accumulation in the liver. Our results are consistent with previous reports from pond and ecological cultures [21]. Overall, aquaculture with submerged macrophytes significantly affected the growth performance and morphology of largemouth bass in the present study, indicating the potential of this model as a reference for farmers.

Effects of sSubmerged Macrophytes on the Nutrient Composition of Largemouth Bass Muscles
Amino acids present a high nutritive value and are important regulators of key metabolic pathways essential for maintenance, growth, feed intake, nutrient utilization, immunity, behavior, and reproduction [38,39]. In the present study, most free and hydrolysable amino acids were more abundant in the EM group; among these, Glu, Asp, Ala, and Arg are well-known as delicious amino acids and contribute significantly to the characteristic flavor of aquatic products [40]. In addition, amino acid content in fish muscles is closely related to their living environment [5,40]. In the present study, levels of 65% amino acids analyzed were significantly and positively correlated with DO and Chlorophyta but significant and negatively correlated with TN and TP (p < 0.05). Furthermore, a significant correlation was noted between amino acids and water environmental factors. Chlorella is highly effective in counteracting fish enteropathy, maintaining a healthy intestine to balance gene expression [41].
Aquatic products are considered to be nutritionally high-quality foods, because they are rich in amino acids and are an excellent source of unsaturated fatty acids, which are beneficial against cardiovascular disease and promote physiological processes [42,43]. In the present study, the content of C18:2, C18:3n3, C20:3n3, and C22:6 (DHA), which can improve human health and nutritional status [44,45], was significantly higher in the EM group. In particular, DHA is beneficial for optimal brain and neuronal development [46] and is an important indicator for evaluating the nutritional value of fatty acids. Moreover, PUFAs (C18:2 and C18:3n3) were significantly and positively correlated with DO and Chlorophyta (p < 0.05). C20:3n3 level was significantly but negatively correlated with cyanobacteria and COD Mn (p < 0.05), while C22:6(DHA) level was significantly but negatively correlated with TN (p < 0.05). In a previous study on channel catfish, long-term exercise was shown to increase bacterial diversity and richness as well as alter the intestinal microbial composition and unsaturated fatty acid and amino acid biosynthesis [47]. Interestingly, water quality (ammonia) affected swimming activity and feeding behavior [48].

Effects of Submerged Macrophytes on Volatile Compounds in Largemouth Bass Muscles
Each food product has a distinct odor imbued by hundreds of volatile organic compounds, and odor change is one of the most sensitive indicators of food quality. Thus, accurately describing the composition of volatile substances can help assess the quality of agri-food products [23,47]. Various flavor components of largemouth bass meat have been documented [5,6]. In the present study, 24 aldehydes accounted for 43.64% of all components and were the most abundant volatile compounds in largemouth bass muscles. These results confirm that different culture environments indeed affect flavor composition. Aldehydes are mainly generated through lipid oxidation and considered to make the greatest contribution to the flavor of meat products because of their higher content and lower odor detection threshold [49,50].
Compared with values in the M group, the levels of aldehydes, such as nonanal-M, nonanal-D, octanal-D, benzaldehyde-M, heptanal-D, 2-methylbutanal-D, and 2-methylbutanal-D, were significantly higher in the EM group. Such differences in aldehydes and other flavor components have been detected in many aquatic products [51]. Benzaldehyde generates pleasant almondy, fruity, and nutty notes [52] and is an important source of the special aroma of crayfish [53]. Meanwhile, the content of hexanal, which was the most abundant aldehyde, was not significantly different between the two groups. Hexanal has previously been identified as an aroma-active compound providing a green note [54].
Typically, alcohols produce a relatively soft odor, similar to the aroma of fruits [55]. Oct-1-en-3-ol is mainly responsible for the green, plant-like aroma and mushroom-like odor and is formed by the oxidation of arachidonic acid by 12-lipoxygenase [56]. Furthermore, ketones are produced through lipid oxidation and generate creamy and fruity notes [57].
2-Butanone was abundant in the muscles of largemouth bass in a recirculatory aquaculture system [6]. In the present study, 2-butanone accounted for approximately 50% of all ketones, being the most abundant one. Thus, 2-butanone appears to be a characteristic volatile compound in largemouth bass.
The higher contents of odor compounds in the muscles of fish from the EM group suggest a more pleasant aroma, which may be another reason for their more delicious meat. The differences in volatile substances between the two culture modes can be explained based on two aspects: (1) improvement of water quality by submerged macrophytes through absorption of excess nitrogen and phosphorus, preventing the outbreak of cyanobacteria and (2) abundance of unsaturated fatty acids in fish muscles. Previous studies have shown that cyanobacterial density in ponds and unsaturated fatty acids in fish muscle are linked to the composition of volatile substances [58,59]. In the present study, cyanobacteria and COD Mn were important biotic and abiotic factors affecting fish, respectively, and these factors were significantly but negatively correlated with 68% of the volatile substances analyzed. Thus, our data suggest that submerged plants play an important role in improving the nutritional composition and characteristic flavor profile of Micropterus salmoides.

Conclusions
In summary, largemouth bass cultured in an ecological pond with submerged macrophytes (Elodea nuttallii) showed optimal growth, with a slender body shape and significantly higher contents of crude protein, total free and hydrolysable amino acids, and ΣPUFA, compared with fish cultured in conventional ponds. Seven aldehydes, nine alcohols, and six ketones were identified as characteristic volatile components in the muscles of largemouth bass cultured in an ecological pond with submerged macrophytes. Variations in the profiles of volatile components between the two groups are closely linked to the diverse water environments caused by the different aquaculture models. Furthermore, cyanobacteria, Chlorophyta, COD Mn , DO, TN, and TP may be the key factors affecting the flavor and nutritional value of largemouth bass. In general, aquaculture with live submerged macrophytes can not only bioremediate the water in situ without producing aquaculture wastewater but also improve the nutritional quality and flavor of aquatic products. Therefore, this is an environmentally friendly and high-value-added ecological aquaculture model worthy of extensive application and popularization.
Author Contributions: Conceptualization, Z.N. and G.X.; methodology, Z.Z. and X.T.; software, Z.Z. and H.Z.; formal analysis, Z.N. and Y.Z.; investigation, Y.S. and J.G.; writing-original draft preparation, Z.Z. and Z.N.; supervision, G.X. and P.X.; project administration, Z.N. and G.X. All authors have read and agreed to the published version of the manuscript.