Effects of High-Density Aquaculture on Grass Carp, Leopard Coral Grouper, and Turbot: Growth Performance and Transcriptome Analysis of Lateral Line

Simple Summary

Stocking density is a critical environmental factor in aquaculture, directly influencing water quality, fish health, yield per unit area and economic returns. However, the effects of stocking density on the physiological responses of the lateral line in different fish species remain unclear. In this study, grass carp, turbot and leopard coral grouper were farmed at different stocking densities. The results indicate that high stocking densities decreased growth and feed efficiency and caused oxidative damage, increasing serum cortisol, malondialdehyde contents, and superoxide dismutase activity in juvenile grass carp (2.04 kg/m³) and turbot (12.61 kg/m³), but did not affect these indicators in juvenile leopard coral grouper (2.33 kg/m³). Meanwhile, high stocking densities decreased the viscerosomatic index and altered the expression of stress-related genes in the lateral line skin across all three species, but did not affect the survival rate and feed intake of the three fish species. This study provides valuable insights for assessing fish health, welfare, and management strategies in intensive aquaculture.

Abstract

The lateral line is a highly differentiated skin sensory organ in fish, but few studies have explored the relationship between stocking density and the physiological responses of the lateral line in different species. In this study, grass carp, turbot and leopard coral grouper were cultured at different stocking densities for 6, 8 or 10 weeks. The results indicate that high stocking densities reduced weight gain and feed efficiency, increased serum cortisol, malondialdehyde contents, and superoxide dismutase activity, and caused oxidative damage in juvenile grass carp (2.04 kg/m³) and turbot (12.61 kg/m³), but did not affect these indicators in juvenile leopard coral grouper (2.33 kg/m³). Meanwhile, high stocking densities did not affect the survival rate and feed intake of the fish, but decreased the viscerosomatic index in all three fish species. In the lateral line skin, high stocking densities upregulated the expression of genes related to glucocorticoid secretion, hypoxia, and oxidative stress in grass carp and turbot, and altered circadian rhythm-related gene expression in leopard coral grouper. The study shows that growth, cortisol level, and oxidative damage can serve as effective indicators for monitoring fish in high-density cultures, and demonstrates that optimal stocking density should be determined based on the farming system, fish species, and developmental stage.

1. Introduction

The global production of aquatic animals reached 94.4 million tons in 2022, and aquaculture serves as an important pillar in the industry, supporting global food security, nutrition supply, and economic development [1]. According to the China fishery statistical yearbook in 2024, China was the world’s largest aquaculture producer and trader, reporting an aquatic product output of 58 million tons, an aquaculture area exceeding 7500 thousand hectares, and an aquaculture output value of over 1.3 trillion yuan in 2023 [2]. Over the past few decades, the aquaculture industry has transformed from extensive to intensive, and stocking density is one of the key factors considered in aquaculture management [3,4,5]. With the rapid development of aquaculture and environmental protection, coastal and inland water areas have become insufficient, leading to high stocking densities becoming a common practice. In intensive or industrial aquaculture, a suitable stocking density can increase yield and economic efficiency per unit volume, but a high density can cause crowding stress and hypoxia, affecting the survival, behavior, feeding, growth, physiological functions, and health of aquatic animals [6,7,8]. For example, in African catfish (Clarias gariepinus, initial weight 13.0 ± 0.2 g) farmed in earthen ponds, a group stocked at 91 g/m³ had a higher final weight, specific growth rate, weight gain and crude protein content than fish in groups with a stocking density of 65 g/m³ or 130 g/m³, but a group with a stocking density of 130 g/m³ had the highest yield per unit area and economic profitability [9]. Similarly, in intensively cultured European perch (Perca fluviatilis L.), the group with a stocking density of 1000 fish/m³ (body weight 25.4 ± 3.9 g) exhibited a significantly higher weight gain, specific growth rate, and condition factor, along with a lower feed conversion rate, than fish in in groups with 600 and 1400 fish/m³, and fin length showed a negative relationship with increasing stocking density [10]. In Largemouth bass (Micropterus salmoides) with an initial weight of 4.50 ± 0.23 g, a medium-density group (0.4 kg/m³) showed a higher weight gain and specific growth rate than fish farmed at a low density (0.2 kg/m³) and a high density (0.6 kg/m³); meanwhile, the high-density group had a higher MDA content and a higher HSP70 gene expression level, as well as lower SOD and CAT activities in the liver, than the low-density group [11]. Conversely, studies on other species demonstrate negative density-dependent growth. In a juvenile blunt snout bream (Megalobrama amblycephala) feeding experiment under different stocking densities (75, 150, 225, 300 and 450 fish/m³), final body weight, weight gain, and specific growth rate showed a significant density-dependent decrease, while serum total protein, triglyceride, cholesterol, and lactate levels showed a significant density-dependent increase [12]. Likewise, in Asian seabass (Lates calcarifer) stocked at densities of 364, 728, 1092, 1456 and 1820 g/m³ (initial weight 5.2 ± 0.1 g), the survival rate, growth performance, feed utilization efficiency, protein efficiency ratio, digestive enzyme activities, and the expression of IGF-1 and GH genes in the liver all declined as stocking density increased [13]. Therefore, identifying the optimal stocking density can help ensure fish welfare, efficient use of space, and maximum production, thereby promoting the healthy development of aquaculture.

The lateral line is a highly differentiated sensory organ of the skin in fish, serving both auditory and tactile functions, and can detect weak water fluctuation and pressure changes [14]. In most fish species, the lateral line comprises thousands of neuromasts distributed across the head, trunk, and tail fin, that are crucial for swimming, feeding, and avoiding danger [15]. In intensive aquaculture with a high stocking density, increased physical contact and collisions among individuals may alter the physiological functions of the lateral line. Nevertheless, few studies have explored the relationship between stocking density and the physiological responses of the lateral line in different fish species. Using high-throughput sequencing technology, transcriptome sequencing (RNA-seq) can measure all RNAs that transcribed by tissues or cells under certain physiological conditions [16]. Transcriptome sequencing has developed rapidly over the past decade, as the inevitable link between genes and proteins, and has become one of the most important tools for molecular biology research [17,18]. In yellow catfish (Pelteobagrus fulvidraco), transcriptome analysis showed that 2750 genes were significantly changed after hypoxia exposure, including upregulation of hif1a, autophagy, and glycolysis/gluconeogenesis pathways in the brain. Meanwhile, the KEGG enrichment of translational efficiency found that lysosome and autophagy were highly enriched [19]. In marine medaka (Oryzias melastigma), hepatic transcriptome analysis under hypoxia found that the endoplasmic reticulum structure and liver metabolism genes were disordered in female fish, and the redox homeostasis and fatty acid metabolism genes were changed in male fish [20]. Therefore, we used transcriptome sequencing to explore the effects of high stocking density on the physiological responses and differentially expressed genes (DEGs) of lateral line skin across multiple fish species.

The main forms of intensive aquaculture include ponds, cages, concrete tanks, and land-based recirculating aquaculture systems. In China, the aquaculture production of marine fish reached 2.06 million tons, and freshwater fish yielded 27.72 million tons in 2023 [1]. Grass carp (Ctenopharyngodon idella), a herbivorous fish primarily cultivated in ponds, is the most widely farmed freshwater fish, with an annual production of 5.90 million tons. Leopard coral grouper (Plectropomus leopardus) and turbot (Scophthalmus maximus) are carnivorous marine fish primarily cultivated in land-based recirculating aquaculture systems, with production yields of 40,000 tons and 50,000 tons, respectively [2]. The grass carp has a docile temperament, while the leopard coral grouper is aggressive and displays cannibalism behavior. The turbot has a flat body, a docile temperament, and is a bottom-dwelling fish with a limited swimming ability. Based on these distinct biological and behavioral traits, we selected these three representative fish species for experiments under different stocking densities. After the experiments, we measured growth performance, serum biochemical indicators, and lateral line skin transcriptomes. Our results clarify the physiological changes of the three fish species under high-density conditions, and provide data support for the healthy development of intensive aquaculture.

2. Materials and Methods

2.1. Experimental Design and Fish Management

This study comprised three independent farming experiments.

Health juvenile grass carp (initial weight 22.7 ± 1.0 g) were purchased and cultured in an outdoor freshwater pond with net cages at the Changde Xidongting management area aquaculture cooperative (Changde, Hunan, China). A total of 2025 experimental grass carp were selected and assigned randomly to 3 treatment groups, including a low-density group (LD-G, 150 fish/tank, 1.03 kg/m³), a middle-density group (MD-G, 225 fish/tank, 1.47 kg/m³), and a high-density group (HD-G, 300 fish/tank, 2.04 kg/m³). The stocking densities for grass carp referred to the local standard of Hebei Province “Technical Specification for Pond Culture of Grass Carp Fry (standard number: DB13/T1028-2009)” [21]. Each group had 3 square net cages, and the length, width and height of the cages were all 1.5 m, and the volume of the cages was 3375 L. Experimental grass carp were hand-fed to apparent satiation twice daily at 8:00 am and 17:00 pm, and the experiment lasted for 6 weeks. The crude protein content of the commercial expanded feed (Tongwei feed, Chengdu, China) exceeds 30%, and the fat content exceeds 5%. During the farming period, water quality parameters were measured every 3 to 5 days based on the conditions of the fish and the weather. The water temperature ranged from 26.0 to 31.0 °C, total ammonia nitrogen content was less than 0.5 mg/L, the dissolved oxygen level was not less than 5.5 mg/L, and pH was between 7.8 and 8.2. Healthy juvenile leopard coral groupers (initial weight 3.9 ± 0.2 g) were purchased and cultured in an indoor flow-through seawater system at the Laizhou Mingbo aquaculture Co., Ltd. (Laizhou, Shandong, China). A total of 360 experimental leopard coral groupers were selected and randomly assigned to 3 treatment groups, including a low-density group (LD-L, 20 fish/tank, 0.79 kg/m³), a middle-density group (MD-L, 40 fish/tank, 1.60 kg/m³), and a high-density group (HD-L, 60 fish/tank, 2.33 kg/m³). The stocking density for leopard coral groupers referred to the local standard of Hainan Province “Technical Specification for Industrial Culture of Leopard Coral Grouper (standard number: DB46/T424-2017)” [22]. Each group had 3 polyethylene tanks, and the volume of the tanks was 100 L. The experimental leopard coral groupers were hand-fed to apparent satiation twice daily at 9:00 am and 16:00 pm, and the experiment lasted for 8 weeks. The crude protein content of the commercial expanded feed (Santong feed, Weifang, China) exceeds 52%, and the fat content exceeds 10%. During the farming period, water quality parameters were measured every 3 to 5 days based on the conditions of the fish and the weather. The water temperature ranged from 27.0 to 30.0 °C, salinity was kept at 25.5 to 29.5 ‰, the total ammonia nitrogen content was less than 0.5 mg/L, the dissolved oxygen level was not less than 6.5 mg/L, and pH was between 7.5 and 8.0.

Healthy turbots (initial weight 96.0 ± 4.5 g) were purchased and cultured in an indoor flow-through seawater system at the Shandong Kehe marine high technology Co., Ltd. (Weihai, Shandong, China). A total of 180 experimental turbot were selected and assigned to 2 treatment groups randomly, including a middle-density group (MD-T, 20 fish/tank, 6.49 kg/m³) and a high-density group (HD-T, 40 fish/tank, 12.61 kg/m³). The stocking density for turbot referred to the local standard of Shandong Province “Technical Specification for Turbot Culture (standard number: DB37/T439-2010)” [23]. Each group had 3 polyethylene tanks, and the volume of the tanks was 300 L. Experimental turbot were hand-fed to apparent satiation twice daily at 8:00 am and 17:00 pm, and the experiment lasted for 10 weeks. The crude protein content of the commercial expanded feed (Saigelin feed, Qingdao, China) exceeds 50%, and the fat content exceeds 10%. During the farming period, water quality parameters were measured every 3 to 5 days based on the conditions of the fish and weather. The water temperature ranged from 15.0 to 20.0 °C, the salinity was kept at 26.0 to 30.5 ‰, the total ammonia nitrogen content was less than 0.5 mg/L, the dissolved oxygen level was not less than 7.2 mg/L, and pH was between 7.5 and 7.7.

2.2. Sample Collection

The experimental grass carp, leopard coral grouper, and turbot were cultured at their respective stocking densities for six, eight, and ten weeks, respectively. Before sampling, all experimental fish were fasted for 12 h, then the group members and total weight per tank were counted. Nine fish per group (three fish per tank) were selected randomly, and anesthetized with MS-222 (10 mg/L) for 2–3 min to collect blood, lateral line skin, and liver tissues. The serum was obtained from blood by centrifuging at 2500 rpm for 10 min, and was stored at −80 °C for biochemical analysis. The muscle and liver samples were collected and immediately frozen with liquid nitrogen. We sampled medium size and representative fish (6–9 fish per group) to measure the body weight and length, and their liver and viscera weights.

2.3. Index Calculation

The formulae for survival rate, weight gain (WG), feed conversion ratio (FCR), feed intake (FI), viscerosomatic index (VSI), hepatosomatic index (HSI) and condition factor (K) were as follows:

Final density (kg/m³) = final fish weight per tank/tank volume;

Survival rate (%) = final fish number/initial fish number × 100;

Weight gain (%) = (final body weight − initial body weight)/initial body weight × 100;

Specific growth rate (%/d) = (ln (final weight) − ln (initial weight))/days of experiment × 100;

Feed conversion ratio = feed intake/(final fish weight − initial fish weight);

Feed intake (%/d) = total dry feed intake/(days of experiment × (initial weight + final weight)/2) × 100;

Hepatosomatic index (%) = liver weight/fish body weight × 100;

Viscerosomatic index (%) = viscera weight/fish body weight × 100;

Condition factor (g/cm³) = body weight/(body length³) × 100.

2.4. Transcriptome Sequencing and Bioinformatic Analysis

Three samples were collected individually from the three fish per tank and were mixed into one sample (three samples per group). Total lateral line skin RNA was extracted using the RNAiso Plus kit (Takara, Japan) according to the manufacturer’s instructions. The RNA solution’s optical density (OD) of 260/280 nm was from 1.9 to 2.0, OD 260/230 ≥ 2.0, RQN ≥ 6.5, 28S/18S ≥ 1.0, respectively, which indicated the high quality of the RNA samples. Then, RNA purification, reverse transcription, library construction, and sequencing (Illumina NovaSeq X Plus platform) were performed by Shanghai Majorbio Bio-pharm Biotechnology Co., Ltd. (Shanghai, China).

The raw paired end reads were trimmed and controlled by fastp with the default parameters, and the GC content, Q30, and Q20 were calculated. Then clean reads of three fishes were separately aligned to their reference genomes from the NCBI database: Ctenopharyngodon Idella (GCF_019924925.1), Plectropomus leopardus (GCF_008729295.1), and Scophthalmus maximus (GCF_022379125.1). Then, the expression level of each transcript was calculated according to the transcripts per million reads (TPM) method, and differential expression gene (DEG) calculation between two different samples was performed using the DESeq2 v 1.3.0. |log2FC| ≥ 1 and FDR < 0.05 (DESeq2) to calculate DEGs between low-density and high-density groups. In addition, KEGG functional enrichment analysis of DEGs was performed by Python scipy software v1.6.0 (p-adjusted < 0.05). All sequencing data were uploaded to the NCBI database, and the GenBank accessions of grass carp, leopard coral grouper, and turbot were PRJNA1303059, PRJNA1303344, PRJNA1303353, respectively. All data were analyzed using the Majorbio bioinformation cloud platform, available online: http://www.i-sanger.com/ (accessed on 2 June 2025).

2.5. Assays of Biochemical Indexes

The serum glucose, triglyceride, total soluble protein, lactate, malondialdehyde (MDA), cortisol levels, superoxide dismutase activity (SOD), and total antioxidant capacity (T-AOC) were measured by using commercial kits from Nanjing Jiancheng Bioengineering Institute (Nanjing, China). The absorbance was read by a microplate reader (Tecan infinite M200, Männedorf, Switzerland), and all measurement steps could refer to relevant kit protocols, available online: http://www.njjcbio.com/ (accessed on 17 June 2025).

2.6. Statistical Analysis

All data are shown as means ± the standard error of the mean (SEM), and were analyzed with SPSS Statistics 21.0 software (IBM corporation, Armonk, NY, USA). All data were tested for normality and homogeneity of variances using the Shapiro–Wilk and Levene’s tests. One-way analysis of variance (ANOVA) was used to evaluate significant differences between the LD, MD, and HD groups in grass carp and leopard coral grouper, followed by a Duncan’s multiple comparison test. The independent-samples t-test was used to evaluate significant differences between the MD and HD groups in turbot, and significant differences were set at p < 0.05. Bar graphs were made using GraphPad prism 8.0 software.

3. Results

3.1. Effects of High-Density Stocking on Growth Performance and Body Indexes in Grass Carp

The initial weight of grass carp was similar across groups, but the final weight was reduced slightly with the increase of stocking density (Figure 1A). The mean initial densities of the LD-G, MD-G, and HD-G groups were 1.03, 1.47, and 2.04 kg/m³, and the corresponding mean final densities were 2.34, 3.46, and 4.52 kg/m³, respectively (Figure 1B). Different stocking densities had no significant effect on the survival rate and feed intake of grass carp (Figure 1C,E). Compared with the LD-G group, the HD-G group had a lower weight gain, specific growth rate, and hepatosomatic index, but had a higher feed conversion ratio (Figure 1D,F, Supplemental Table S1). Both the MD-G and HD-G groups had a lower viscerosomatic index and condition factor than the LD-G group (Figure 1G,H), but there were no significant differences between these indicators. These data suggest that an initial stocking density of 2.04 kg/m³ inhibits the growth of juvenile grass carp.

3.2. Effects of High-Density Stocking on Serum Biochemical Indicators in Grass Carp

Compared with the LD-G group, the serum glucose content was significantly reduced in both the MD-G and HD-G groups (Figure 2A). Different stocking densities did not affect the triglyceride and lactic acid levels in the serum (Figure 2C,D). The serum cortisol and malondialdehyde contents were increased with the increase of stocking density (Figure 2E,F). The MD-G and HD-G groups all had higher total protein content, total antioxidant capacity, and superoxide dismutase activity than the LD-G group, and a significant difference was found in the HD-G group (Figure 2B,G,H). These results suggest that a high stocking density lead to oxidative stress in juvenile grass carp.

3.3. Effects of High-Density Stocking on Transcriptome of Lateral Line in Grass Carp

As shown in

Table 1. Quality statistics of filtered reads from the lateral line transcriptome of grass carp.

Sample	Read Sum	Base Sum	GC (%)	Q20 (%)	Q30 (%)
LD-G1	46,813,438	7,068,829,138	47.78	98.58	95.71
LD-G2	45,740,416	6,906,802,816	47.13	98.57	95.66
LD-G3	42,828,174	6,467,054,274	48.41	98.56	95.59
HD-G1	41,569,364	6,276,973,964	47.58	98.63	95.88
HD-G2	48,903,346	7,384,405,246	47.93	98.58	95.7
HD-G3	47,805,848	7,218,683,048	47.19	98.58	95.72

, RNA-seq was performed on lateral line skin in the LD-G and HD-G groups. After quality control, GC content was around 47–48%, and the percentage of Q30 bases in each sample was no less than 95%, indicating high-quality transcriptome data suitable for subsequent analysis. A total of 114 DEGs were identified, including 21 upregulated and 93 downregulated genes (Figure 3A). KEGG enrichment analysis showed that arginine and proline metabolism, mTOR signaling, autophagy, alanine, aspartate, glutamate metabolism, and glycerophospholipid metabolism signaling pathways were significantly enriched (Figure 3B). The heatmap of DEGs showed that slc3a2b (solute carrier family 3 member 2b), irs2b (insulin receptor substrate 2b), sgk2b (serum/glucocorticoid regulated kinase 2b), mylk4a (myosin light chain kinase family, member 4a), mylk4b (myosin light chain kinase family, member 4b), hif1al (hypoxia inducible factor 1 subunit alpha, like), and ddit4 (DNA damage inducible transcript 4) gene expression were increased, while oat (ornithine aminotransferase), cndp1 (carnosine dipeptidase 1), wnt2bb (wingless-type MMTV integration site family, member 2Bb), asip2b (agouti signaling protein 2b), and chst7 (carbohydrate sulfotransferase 7) genes expression were reduced (Figure 3C).

3.4. Effects of High-Density Stocking on Growth Performance and Body Indexes in Leopard Coral Groupers

After 8 weeks of farming, the different stocking densities had no significant effect on the final weight of leopard coral groupers (Figure 4A). The mean initial densities of the LD-L, MD-L, and HD-L groups were 0.79, 1.60, and 2.33 kg/m³, with the mean final densities reaching 4.03, 8.07, and 11.67 kg/m³, respectively (Figure 4B). The different stocking densities had no significant effect on the survival rate, weight gain, specific growth rate, feed intake, feed conversion ratio, and hepatosomatic index of leopard coral groupers (Figure 4C–F, Supplemental Table S2). Additionally, the HD-L group had a lower viscerosomatic index, and the MD-L group had a lower condition factor than the LD-L group (Figure 4G,H). These results show that the initial stocking densities of 0.79–2.33 kg/m³ did not affect the growth of juvenile leopard coral groupers.

3.5. Effects of High-Density Stocking on Serum Biochemical Indicators in Leopard Coral Groupers

In leopard coral groupers, different stocking densities had no significant effect on glucose, lactic acid, cortisol, and malondialdehyde contents in the serum (Figure 5A,D–F). The serum total protein content increased with the increase of stocking density, and a significant difference was found for the HD-L group (Figure 5B). Compared with the LD-L group, the serum triglyceride level was significantly reduced in the MD-L group (Figure 5C), while the total antioxidant and superoxide dismutase activities were slightly decreased in the HD-G group (Figure 5G,H). These data indicate that high stocking density did not cause oxidative stress and damage in leopard coral groupers.

3.6. Effects of High-Density Stocking on Transcriptome of Lateral Line in Leopard Coral Groupers

As shown in

Table 2. Quality statistics of filtered reads from the lateral line transcriptome data of leopard coral groupers.

Sample	Read Sum	Base Sum	GC (%)	Q20 (%)	Q30 (%)
LD-L1	42,082,068	6,354,392,268	46.14	98.39	95.29
LD-L2	42,796,646	6,462,293,546	47.57	98.44	95.37
LD-L3	43,051,198	6,500,730,898	46.73	98.4	95.31
HD-L1	47,006,932	7,098,046,732	47.2	98.41	95.35
HD-L2	42,505,422	6,418,318,722	47.7	98.46	95.44
HD-L3	46,593,760	7,035,657,760	47.01	98.4	95.33

, RNA-seq was performed on lateral line skin in the LD-L and HD-L groups. After quality control, GC content was around 46–47%, and the percentage of Q30 bases in each sample was no less than 95%, indicating high-quality transcriptome data suitable for subsequent analysis. A total of 175 DEGs were identified, including 56 upregulated and 119 downregulated genes (Figure 6A). KEGG enrichment analysis showed that circadian rhythm, citrate cycle, PPAR family, and cytokine pathways were significantly enriched (Figure 6B). The heatmap of DEGs showed that mx (mx dynamin like GTPase), mx1 (mx dynamin like GTPase 1), il1rl1 (interleukin 1 receptor like 1), per3 (period circadian regulator 3), per1b (period circadian regulator 1b), tef (thyrotrophic embryonic factor), and adrb1 (adrenoceptor beta 1) gene expression rose, while cdkn1d (cyclin dependent kinase inhibitor 1d), clocka (circadian locomotor output cycles kaput a), tgfβ3 (transforming growth factor beta 3), aco2 (aconitase 2), and ogdh (oxoglutarate dehydrogenase) gene expression decreased (Figure 6C).

3.7. Effects of High-Density Stocking on Growth Performance and Body Indexes in Turbot

Compared with the MD-T group, the final weight of turbot was significantly decreased in the HD-T group (Figure 7A). The mean initial densities of the MD-T and HD-T groups were 6.49 and 12.61 kg/m³, with the mean final densities reaching 14.59 and 26.09 kg/m³, respectively (Figure 7B). Different stocking densities had no significant effect on the survival rate, feed intake, and condition factor of turbot (Figure 7C,E,H). Compared with the MD-T group, the HD-T group had significantly lower weight gain, specific growth rate, hepatosomatic index, and viscerosomatic index (Figure 7D,G, Supplemental Table S3). Meanwhile, the HD-T group had a higher feed conversion ratio, but there was no significant difference (Figure 7F). These data show that high-density stocking at 12.61 kg/m³ reduced the growth of turbot.

3.8. Effects of High-Density Stocking on Serum Biochemical Indicators in Turbot

Compared with the MD-T group, the HD-T treatment did not affect the glucose content and total antioxidant capacity in the serum (Figure 8A,G). The HD-T group had a significantly lower total protein content in the serum (Figure 8B), but had higher triglyceride and lactic acid contents than the MD-T group (Figure 8C,D). In addition, the serum cortisol, malondialdehyde contents, and superoxide dismutase activity were increased in the HD-T group compared to the MD-T group (Figure 8E,F,H). These data show that a high stocking density causes oxidative stress in turbot.

3.9. Effects of High-Density Stocking on Transcriptome of Lateral Line in Turbot

As shown in

Table 3. Quality statistics of filtered reads from the lateral line transcriptome of turbot.

Sample	Read Sum	Base Sum	GC (%)	Q20 (%)	Q30 (%)
MD-T1	47,180,048	7,124,187,248	49.42	98.66	95.88
MD-T2	44,941,552	6,786,174,352	48.17	98.52	95.51
MD-T3	45,130,388	6,814,688,588	48.71	98.51	95.48
HD-T1	40,966,554	6,185,949,654	48.76	98.49	95.41
HD-T2	45,978,284	6,942,720,884	48.99	98.59	95.68
HD-T3	41,744,192	6,303,372,992	48.04	98.52	95.51

, RNA-seq was performed on lateral line skin in the MD-T and HD-T groups. After quality control, the GC content was around 48–49%, and the percentage of Q30 bases in each sample was no less than 95%, indicating high-quality transcriptome data suitable for subsequent analysis. A total of 242 DEGs were identified, including 66 upregulated and 176 downregulated genes (Figure 9A). KEGG enrichment analysis showed that African trypanosomiasis, malaria, hematopoietic cell lineage, PPAR family, and tyrosine metabolism pathways were significantly enriched (Figure 9B). The heatmap of DEGs showed that apoa1 (apolipoprotein a1), pparβ (Peroxisome Proliferator-Activated Receptor Beta), fabp7 (fatty acid binding protein 7), il4i1 (interleukin 4 induced 1), hba1 (hemoglobin subunit alpha-1), and hbz (hemoglobin subunit zeta) gene expression was elevated, but that il1β (interleukin 1 beta), socs3 (suppressor of cytokine signaling 3), hspa1a (heat shock protein family a members 1a), hspa1b (heat shock protein family a members 1b), fga (fibrinogen alpha chain), and gabrb1 (gamma-aminobutyric acid type a receptor subunit beta1) were downregulated (Figure 9C).

4. Discussion

A high stocking density represents an intensive and highly efficient model of animal husbandry. In 2023, the output value of marine aquaculture in China reached 488.55 billion yuan, and that of freshwater aquaculture totaled 817.79 billion yuan, reflecting a continued positive trend in fishery production and economic performance [2]. Stocking density is the most important production factor in aquaculture, and affects water quality, growth performance, and protective barriers (skin, intestine, and gills) against invading pathogens [6,7]. Some studies found that appropriately increasing stocking density does not affect growth and survival. For example, in rainbow trout (Oncorhynchus mykiss) cultured at two stocking densities (initial densities 12 kg/m³ and 17 kg/m³, final densities 21 kg/m³ and 30 kg/m³), final body weight, specific growth rate, protein efficiency rate, feed conversion rate, and swimming activity showed no significant differences between two stocking densities during the experiment duration [24]. Similarly, in European sea bass (Dicentrarchus labrax) with an initial weight of 72 ± 4 g, the high density group (36 kg/m³) did not affect growth performance and feed intake, but showed lower energy maintenance requirements and higher cortisol levels in the blood than the low density group (5.5 kg/m³) [25]. Conversely, numerous studies found that an excessively high stocking density can reduce survival rate, growth, and the feeding efficiency of fish. In minor carp (Labeo bata) stocked at 50, 75, and 100 fish/m³ (initial weight 6.18 ± 1.32 g), the survival rate, specific growth rate, feed efficiency, and protein efficiency ratio were highest in fish reared at 50 fish/m³ compared to fish reared at 75 fish/m³ or 100 fish/m³ [26]. In a Nile tilapia feeding experiment with stocking densities of 18.75, 37.50, 56.25, and 75.00 fish/m³ (initial weight 133.91 g), higher final body weight and weight gain were observed in the lowest stocking density group; meanwhile, higher gill branch lesions and lower ash content were observed in the highest density group [27]. In channel catfish (Ictalurus punctatus) stocked at 50, 150, and 300 fish/m³, weight gain, feed efficiency, muscle fat content, and muscle fiber diameter decreased significantly with increasing stocking density, whereas serum glucose, triglyceride, total cholesterol levels, aspartate aminotransferase, and alanine aminotransferase activities increased significantly [28]. Similar trends were observed in Japanese flounder (Paralichthys olivaceus) stocked at 500, 1000, 1500, 2000, 2500 ind/m³ (initial weight 1.27 ± 0.04 g), where final body weight, specific growth rate, feed efficiency, and protease activity decreased significantly as stocking density increased [29]. In dourado native fish (Salminus brasiliensis) stocked at densities of 0.24, 1.17, and 1.70 g/m³, weight gain decreased with increasing density, and enzyme activities of GST, GPX, and CAT in the liver, along with Na⁺ and Cl⁻ concentrations in the gills, increased with increasing density [30]. In the present study, high stocking densities reduced weight gain and increased feed conversion ratio in juvenile grass carp (2.04 kg/m³) and turbot (12.61 kg/m³), but did not significantly affect these indicators in juvenile leopard coral groupers (2.33 kg/m³). Meanwhile, high stocking densities had no significant effect on survival rate and feed intake across all three species, and the viscerosomatic index decreased consistently with increasing density. In our study, high-density farming reduced the growth of grass carp slightly but did not affect the growth of leopard coral groupers. The differential growth responses among the three species may be attributed to several factors. Grass carp were farmed in static pond cages while leopard coral groupers were farmed in a flow-through system with higher dissolved oxygen levels. Another possible reason is that leopard coral groupers had a lower initial weight (3.9 ± 0.2 g) than grass carp (22.7 ± 1.0 g) and turbot (96.0 ± 4.5 g). Additionally, as a coral reef-associated species, the leopard coral grouper may have developed a greater tolerance to high stocking densities through natural adaptation to physical contact with reefs during swimming. However, turbot showed a significant growth reduction at high densities, likely due to their flat body shape, which caused overlapping and physical interference during culture. In Asian swamp eel (Monopterus albus) reared at stocking densities of 8.4, 15.0, and 21.7 kg/m³ (initial weight 16.67 ± 3.48 g), the medium density group had a higher survival rate, growth, and nutrient accumulation [31]. Similarly, in largemouth bass (Micropterus salmoides) stocked at 90.91, 113.63, and 136.36 ind./m³ (initial weight 8.25 ± 0.51 g), the middle density group achieved maximal final body weight, weight gain, specific growth rate, and yield after 300 days of culture [32]. These results show that an appropriate stocking density could increase fish production. In Asian seabass (Lates calcarifer), three stocking densities (0.03, 0.04 and 0.05 ind/L) did not affect the body weight and body length, but reduced survival rate. However, fish cultured in recirculating aquaculture systems had higher weigh gain, specific growth rate, lower feed conversion ratio, and cannibalism than those in static water systems [33]. Intensive aquaculture can cause serious negative impacts to water environments and aquatic animals, and RAS is a sustainable and efficient practice that promotes the production and health status of fish. The above data indicate that an appropriate stocking density can increase fish production, while an excessively high density can inhibit fish growth, and that growth performance is also influenced by farming system, fish species, and body size.

In the stress responses of animals, cortisol, glucose, and oxidative damage serve as key biochemical parameters for assessing changes to physiological state, reflecting the stress process from neuroendocrine regulation to metabolic mobilization and potential pathological damage [34]. As the primary stress hormone, cortisol is rapidly released into the blood upon exposure to stressors (such as fright, crowding, or water deterioration) through the activation of the hypothalamic–pituitary–adrenal (HPA) axis [35]. Cortisol can regulate metabolism, inflammatory, immune, and stress responses, including the promotion of glycogen breakdown, gluconeogenesis, and blood glucose levels to meet the heightened energy and survival demands of the organism [36]. In rainbow trout (Oncorhynchus mykiss) cultured at two stocking densities (initial densities 12 kg/m³ and 17 kg/m³, final densities 21 kg/m³ and 30 kg/m³), no significant differences were observed in final body weight, specific growth rate, protein efficiency rate, feed conversion rate, swimming activity, and serum stress indicators (cortisol, glucose, lactate levels, hematocrit, red blood cell count, and lysozyme activity) [24]. Conversely, in common carp (Cyprinus carpio) stocked at 6 kg/m³ and 12 kg/m³ (initial weight 8.41 ± 0.44 g), serum catalase, superoxide dismutase, glutathione peroxidase activities, and malondialdehyde level were increased along with the elevation of stocking density, while lipoprotein, cholesterol, triglyceride, total protein, and albumin levels were decreased [37]. Similarly, largemouth bass at a high stocking density (136.36 ind./m³, initial weight 8.25 ± 0.51 g) showed higher glucose, triglyceride, total cholesterol, total bilirubin, malondialdehyde levels, alanine transaminase, and aspartate transaminase activities in the serum, a lower total protein level in the serum, amylase, lipase, and trypsin activities in the intestine, superoxide dismutase and catalase activities in the liver, and crude lipid and saturated fatty acid in the muscle than that in fish in the low stocking density group (90.91 ind./m³) [32]. Notably, in another largemouth bass study, a high stocking density (0.6 kg/m³) did not alter the serum cortisol, glucose, and triglyceride levels during 30–90 days of cultivation, but these indexes were increased significantly in the high-density group at 120 days, underscoring the influence of culture duration and developmental stage on stress response [11]. Additional rainbow trout studies reported, for the high density group (25 kg/m³), increased feed conversion ratio, red blood cell, hemoglobin levels, and moisture content of whole fish, but decreased crude protein and lipid contents of whole fish compared with the low density group (15 kg/m³) [38]. The growth, serum complement activity, osmolality, and globulin levels were decreased by increasing the stocking density from 20 kg/m³ to 80 kg/m³, but albumin and malondialdehyde levels, lysozyme, catalase, glutathione peroxidase, and superoxide dismutase activities were raised [39]. In the present study, high-density stocking significantly increased serum cortisol, malondialdehyde, oxidative stress, total protein contents, total antioxidant, and superoxide dismutase activities, but reduced serum glucose content in grass carp. In turbot, high-density stocking increased serum triglyceride, lactic acid, cortisol, malondialdehyde contents, and superoxide dismutase activity, but decreased total protein content. However, in leopard coral groupers, high-density stocking increased serum total protein content, and decreased serum triglyceride level, total antioxidant capacity, and superoxide dismutase activity, but did not affect the glucose, lactic acid, cortisol, and malondialdehyde contents in the serum. Our results indicate that grass carp and turbot exhibited significant stress responses under high stocking densities, but the leopard coral groupers showed fewer stress responses. High stocking density is a cause of chronic stress associated with elevated cortisol levels, which can lead to metabolic disorders and oxidative damage. High cortisol concentrations suppress growth, increase the generation of ROS, and disrupt cellular redox balance. At this stage, oxidative damage indicators such as malondialdehyde (a product of lipid peroxidation) increase, while the activities of antioxidant enzymes like superoxide dismutase and catalase exhibit an initial compensatory rise followed by a decline due to depletion.

As is widely acknowledged, an increased stocking density can decrease dissolved oxygen content and change other water quality indicators such as ammonia, pH, and carbon dioxide [31]. However, it remains unclear whether crowded environments and water quality changes affect lateral line skin in fish. In Atlantic salmon (Salmosalar L.) stocked at 10, 30, 50, and 70 kg/m³, high stocking densities induced a decline in water quality and chronic stress in the fish, including gut barrier and immune system damage, as well as increased neutrophil infiltration [40]. Similarly, in rainbow trout cultures at different stocking densities (10, 40, and 80 kg/m³), an elevated density and crowding increased HSP70 gene expression in the head kidney and serum stress responses including high adrenocorticotropic hormone, cortisol, and lactate levels, while reducing innate immune responses (serum total antioxidant, complement activity, IgM content, TNF-1α, IL-8, and IFN-γ1 genes expression) and health status [41]. In a feeding experiment on grass carp cultured at different stocking densities (0.9 kg/m², 2.97 kg/m² and 5.9 kg/m²), the transcript levels of inflammatory cytokine genes (il1β and tnfα), as well as the apoptosis-related genes (caspase 8, fasl, and caspase3), were significantly increased in the high-density group, whereas serum lysozyme, acid phosphatase, and alkaline phosphatase activities were significantly decreased [42]. The transcriptome results of lateral line skin showed that the expression of insulin signaling, glucose metabolism (irs2b), glucocorticoid secretion (sgk2b), hypoxia stress (hif1al), and DNA damage (ddit4)-related genes were increased in the high stocking density cultured grass carp, and the changes are consistent with the corresponding biochemical indicators. These data illustrate that high stocking densities lead to hypoxia and metabolic disorders in grass carp, as well as increased cortisol secretion and oxidative damage. Autophagy serves as a protective adaptation by degrading damaged organelles and proteins, clearing oxidatively damaged components, and providing cells with emergency energy and raw materials [43]. The mTOR and autophagy pathways jointly mediate the transition of the organism from a “growth mode” to a “survival mode,” reflecting the adaptive strategy of grass carp under high-density stress. In turbot, a high stocking density also led to hypoxia and the expression of hemoglobin synthesis (hba1and hbz) genes, and especially increased expression levels of lipid metabolism-related genes (apoa1, pparβ, and fabp7). As turbot primarily store fat in the subcutaneous layer and liver, their bodies overlap under high-density farming, subjecting the skin to significant pressure, friction, and damage. The above results suggest that fatty acid oxidation and lipid transport play crucial roles in cellular energy demands and membrane repair under chronic stress [44]. Meanwhile, high stocking density inhibited inflammatory responses (il1β and socs3) and molecular chaperone (hspa1a, hspa1b, and gabrb1)-related gene expression, thereby suppressing the skin’s inflammatory response, mucosal defense, and tissue repair capacity. In this study, the water quality parameters showed no difference between the normal and high stocking density groups, indicating that the observed hypoxia-related responses likely represent secondary physiological consequences induced by crowding stress rather than direct water quality effects. In leopard coral groupers, high stocking density significantly changed the expression of circadian rhythm-related genes (per3, per1b, tef, cdkn1d, and clocka), which indicates that high stocking density caused comparatively fewer stress responses in this species. The skin of the leopard coral grouper is red, and it is highly sensitive to light intensity. Insufficient light exposure induced by a high-density culture can directly disrupt the circadian system, thereby affecting the expression of clock genes [45]. In the leopard coral grouper, circadian rhythm disruption may represent merely the initial response to high density-related stress, with potential subsequent effects on growth performance and health status. Contrary to expectations, a large number of DEGs were not identified in the lateral line skin across all three species under crowding conditions. This may demonstrate that the lateral line, as a sensory organ, exhibits a highly specific response to high-density stress.

In summary, the increase in serum cortisol serves as an initial biomarker and regulatory center to crowding response stress in fish. Subsequent changes in nutrient levels reflect the state of metabolism efficiency and energy supply, while oxidative damage indicators and growth performance reveal the direct harms caused by stress. The above indicates that fish redirect physiological resources from growth and energy production toward immune defense and stress adaptation. Together, these three parameters form a continuous monitoring framework, spanning stress perception, physiological regulation, and direct damage, and providing critical insights for assessing fish health, welfare, and aquaculture management strategies.

5. Conclusions

Stocking density is a critical production factor in aquaculture. The study shows that high stocking densities reduced weight gain and feed efficiency, and caused oxidative damage, including increases in serum cortisol, malondialdehyde contents, and superoxide dismutase activity in juvenile grass carp (2.04 kg/m³) and turbot (12.61 kg/m³), but did not affect these indicators in juvenile leopard coral groupers (2.33 kg/m³). Meanwhile, different stocking densities did not affect the survival rate and feed intake in the three fish species, but decreased the viscerosomatic index and changed stress-related gene expression in the lateral line skin of all three. The results provide important references for the healthy farming of fish and the management of intensive aquaculture.