2.1. Feed Preparation

Seven isonitrogenous (38.23%) and isolipidic (8.50%) diets were prepared. The control diet (CON) contained 40% FM. SBM was fermented by Bacillus subtilis according to a published method [23]. In brief, after extended incubation, B. subtilis was inoculated into the sterilized SBM. The SBM was immediately placed in a shaker and fermented at 37°C, 100 rpm, and 80% humidity for 24 h. The FSBM was then placed in an oven and dried at 50°C for 6–7 h to maintain 8%–10% humidity. Finally, FSBM was crushed and stored at −20°C until use. FSBM was used to replace 50%, 75%, and 100% of FM, which were named 50FSBM, 75FSBM, and 100FSBM, respectively. The amino acid composition of FSBM and FM is detailed in Tables S1 and S2, respectively. To evaluate the effect of glycine on improving the utilization of plant-based protein, 1% glycine was added to the 50FSBM, 75FSBM, and 100FSBM groups, which were named 50FSBMG, 75FSBMG, and 100FSBMG, respectively. Alanine was used as the isonitrogenous control to balance the total nitrogen content in different experimental diets. The experimental diets were formulated with a previously published method [27]. In brief, the feed ingredients were crushed and sifted through a 60-mesh sieve. The ingredients were then weighed and combined with fish oil and distilled water. The mixture was subsequently formed into sinking pellets with diameters of 0.8 and 1.0 mm by a twin-screw extruder. The wet feed was then subjected to drying at 45°C for 24 h, followed by storage at −20°C for future use. The feed formulation and composition are presented in Table 1, and the composition of amino acids in the experimental feeds is shown in Table 2.

2.2. Shrimp Rearing and Sample Collection

Whiteleg shrimp for aquaculture were purchased from Zhejiang Yize Aquaculture Co. Ltd. (Taizhou, Zhejiang, China). After 14 days of acclimatization to the experimental environment, 420 healthy shrimp (initial weight: 1.1 g) were randomly distributed into 21 tanks (water volume: 300 L; size: 100 cm × 60 cm × 80 cm), with 20 shrimp per tank. Seven experimental diets were distributed in the experimental shrimp tanks, with three replicates for each treatment. The feeding frequency of shrimp was maintained at four times (at 06:00, 11:00, 16:00, and 22:00) a day and lasted for 8 weeks. The daily feeding consumption was determined by the body weight with a ratio of 5%–8% and adjusted every 2 weeks. The water quality was maintained at optimal conditions for shrimp feeding and growth (temperature: 24–26°C; nitrite < 0.1 mg/L; dissolved oxygen level ≥ 6 mg/L; total ammonia nitrogen <0.5 mg/L; pH: 7.8–8.3; salinity: 29–31‰).

At the end of the animal experiment, all shrimp were subjected to a 24 h fast. All shrimp in each tank were subsequently counted and weighed to analyze weight gain rate (WGR), SGR, feed efficiency ratio (FER), and survival rate (SR). In addition, six shrimp from each tank were dissected, and their hepatopancreas were weighed to determine the hepatopancreas index (HSI). The muscle of the shrimp was removed; one part was preserved in paraformaldehyde (4%), and the other part was fixed in glutaraldehyde. The samples of hepatopancreas and muscle were frozen in liquid nitrogen and stored at -80°C until use.

The following parameters were calculated:

WGR (%) = 100 × (final body wet weight − initial body wet weight)/initial body wet weight;

SGR (%/day) = 100 × (Ln (final body wet weight) − Ln (initial body wet weight))/duration of the experiment (in days);

HSI (%) = 100 × hepatopancreas wet weight/final body wet weight;

FER (%) = 100 × body wet weight gain/dry feed intake.

2.3. Proximate Composition and Amino Acid Analysis

The moisture, crude protein, crude lipid, and ash contents of shrimp and feed were measured via methods provided by the Association of Official Analytical Chemists (AOAC) [28]. The feed and shrimp samples were dehydrated to constant weight with a lyophilizer (Freeze Dryer LL1500, Thermo Fisher Scientific) for the moisture analysis. Crude protein content was determined via a Kjeldahl apparatus (K355/K437, Buchi). In brief, samples were coheated with strong sulfuric acid. Subsequently, sodium hydroxide (NaOH) was added for distillation, and the amount of nitrogen collected by boric acid was determined by titration with hydrochloric acid. Crude lipid content was quantified by a Soxhlet apparatus. At 55°C water bath, the samples were placed in lipid packets and extracted by continuous refluxing petroleum ether for 12 h. The weight of the dried lipid packet was used to calculate crude lipid content. Ash content was measured by burning samples in muffle at 550°C for 12 h.

The contents of amino acids in the diets, whole body, and muscle were measured using previously described methods [26, 29]. Briefly, the sample was placed in the ampoule, followed by adding 10 mL of hydrochloric acid (6 mol/L). The ampoule was filled with nitrogen, sealed, and then placed in an oven, followed by hydrolysis at 110°C for 24 h. The hydrolysate was centrifuged at 10,000 rpm for 10 min. Then, the supernatant was analyzed using a liquid chromatograph (1260 Infinity II, Agilent).

2.4. Histological Analysis of Hepatopancreas and Muscle

Hematoxylin and eosin (H&E) staining was applied to muscle and hepatopancreas tissues for histological analysis. In brief, tissues preserved in 4% paraformaldehyde were transferred to 70% alcohol overnight, dehydrated through a series of ascending 80%, 90%, and 100% ethanol solutions, and embedded in paraffin. Sections (~5–8 μm each) were obtained by a microtome and stained with H&E. The stained sections were then photographed under a microscope (Olympus BX53, Japan). For transmission electron microscopy of longitudinal sections of muscle, the muscle tissue was fixed with glutaraldehyde and sent to Servicebio Co. Ltd. (Wuhan, Hubei, China).

2.5. Antioxidant Indicators

The antioxidant indices in the muscle and hepatopancreas, including reduced glutathione (GSH), total protein (TP), superoxide dismutase (SOD), and total antioxidant capacity (T-AOC) were measured with commercial test kits (Jiancheng Bioengineering Inc., Nanjing, China). Muscle and hepatopancreas samples were homogenized in a 1:9 (w:v) saline solution and centrifuged for 10 min at 2500 rpm.

For TP detection, the reagents in the commercial kits were mixed with samples and incubated at 37°C for 20 min. Subsequently, the mixture was analyzed with a microplate reader at 562 nm. For T-AOC analysis, samples were combined with reagents and incubated at 25°C for 6 min, after which the absorbance of the mixture was measured with a microplate reader at 405 nm. For SOD activity measurement, before testing, samples were diluted to different concentrations to adjust the inhibition rate of SOD, which remained within the range of 40%–60%. The diluted samples were mixed with reagents and scanned at 450 nm using a microplate reader. For the analysis of GSH content, after the standard in the kit was diluted into 20 μmol/mL GSH standard solution, the solution was mixed with the samples. The mixture was then stalled for 5 min, and the absorbance was then measured at 405 nm using a microplate reader. For MDA measurement, the samples and reagents were mixed in a 95°C water bath. After 40 min of incubation, the mixture was centrifuged for 10 min at 3500 rpm, and the supernatant was measured with a microplate reader at 532 nm.

2.6. RNA Extraction, cDNA Synthesis, and Quantitative Real-Time PCR

Total RNA in muscle was extracted using TRIzol Reagent (Vazyme, China). In brief, after the samples were homogenized with 1 mL of TRIzol, the homogenate was centrifuged for 10 min at 12,000 x g. The supernatant was collected, mixed with 200 μL of chloroform, and shaken vigorously for 30 s. Then, the mixture was incubated at 4°C for 3 min and centrifuged for 15 min at 12,000 x g. The supernatant was collected, and 400 μL of isopropanol was added for centrifugation (12,000 x g, 10 min). The precipitate was washed with 75% ethanol twice and solubilized by the addition of nuclease-free water. After the concentration of extracted RNA was measured by Nanodrop 2000, the extracted RNA was reverse transcribed to cDNA using HiScript III RT SuperMix for qPCR (+gDNA wiper) (Vazyme, China) through a two-step process. RT–qPCR was performed using SYBR qPCR Master Mix (Vazyme). In brief, the sample and the reagents in the kit were mixed according to the instructions. The mixture was kept at 95°C for 3 min and subjected to 39 cycles. The cycle conditions were 95°C for 10 s, 58°C for 10 s, and 72°C for 20 s. Primers were designed according to the public NCBI sequences and made in Sangon Biotech. Co., Ltd. (Shanghai, China). The details are presented in Table 3. In this study, β-actin was selected as the reference gene because the expression of β-actin was the most stable among the different treatments. The gene mRNA levels were quantified using the comparative cycle threshold (CT) approach (2^−∆∆ct method) [30].

2.7. Statistical Analysis

The data of this study are shown as the means ± standard deviation (SD). After analyzing for normality and homoscedasticity, one-way analysis of variance (ANOVA) and Tukey’s test were used to assess the differences between the various groups in SPSS 27.0 software. A statistically significant difference was defined as p < 0.05.

3.1. Dietary Glycine Supplementation Alleviated FSBM Substitution-Induced Growth Inhibition

After an 8-week feeding trial, the growth parameters of whiteleg shrimp are presented in Table 4. The SRs of all groups were greater than 90% with no significant difference (p > 0.05). With the rise of the FSBM substitution ratio, final body weight, WGR, SGR, and FER significantly decreased (p < 0.05). Interestingly, compared with the 50FSBM group, 1% dietary glycine supplementation significantly enhanced the growth performance of shrimp (p < 0.05). Moreover, there were no significant differences in growth parameters between the 50FSBMG group and the control group. Although the final body weight, WGR, and SGR in the 75FSBMG group were significantly higher than those in the 75FSBM group (p < 0.05), they were still significantly lower than those in the control group. However, dietary supplementation with 1% glycine supplementation did not substantially improve the growth performance compared with that of the 100FSBM group (p > 0.05). Glycine supplementation significantly blocked FSBM substitution-induced increase in the HSI of shrimp (p < 0.05). There was no significant effect on the body composition of the whole shrimp body among all the groups (p > 0.05, Table 5).

3.2. Dietary Glycine Supplementation Altered the Amino Acid Profile of Whole Shrimp and Muscle

The amino acid composition of feed in each group accurately reflected the amino acid differences between FM and FSBM. FSBM substitution significantly decreased the levels of methionine, lysine, total essential amino acids, and glycine but increased the glutamate content (p < 0.05, Table 2). Glycine supplementation significantly increased the glycine levels in the 50FSBMG, 75FSBMG, and 100FSBMG groups (p < 0.05, Table 2). In the whole body, with the rise of FSBM substitution ratio, the contents of branched-chain amino acids (BCAAs), including valine, isoleucine, and leucine, tended to decrease, but the difference was not significant (p > 0.05, Table 6). Compared with the 75FSBM and 100FSBM groups, dietary glycine supplementation increased the levels of isoleucine and leucine, but the difference was not substantial (p > 0.05, Table 6). The total essential amino acid content in the whole shrimp body was not affected by the feed composition (p > 0.05, Table 6). However, the glycine content in 75FSBMG and 100FSBMG groups was significantly higher than that in the control group (p < 0.05, Table 6). In the muscle, with the rise of FSBM substitution ratio, the contents of valine, phenylalanine, isoleucine, leucine, lysine, total essential amino acids, and glycine considerably decreased (p < 0.05). Glycine content in the 50FSBMG group was significantly higher than that in the 50FSBM group (p < 0.05). Correspondingly, the 100FSBMG group significantly increased the glycine content compared with the 100FSBM group (p < 0.05, Table 7).

3.3. Dietary Glycine Supplementation Enhanced the Structure of Hepatopancreas and Muscle Under FSBM Substitution

As the main center of metabolism and digestion in shrimp, hepatopancreas function depends on healthy structures, which are also affected by nutritional stress. In the present study, the hepatopancreas in the control group was a complex tubular gland composed of multiple glandular ducts, closely adjacent to each other, with lumens in the shape of quadrilaterals. The secretory cells and absorbent cells in the tube wall had a regular and clearly visible shape (Figure 1). However, in the FSBM substitution group, air bubbles were occasionally observed in the hepatopancreas. In addition, the atrophy of the gland ducts was severe and the distance between the gland ducts increased. The gland ducts were no longer adjacent to each other. In the 100FSBM group, the epithelium of the ducts detached from the basement membrane. On the basis of 50% and 75% substitution, adding glycine improved the structure of hepatopancreas. However, glycine could not reverse the damage caused by 100% substitution (Figure 1).

FSBM substitution and glycine supplementation also affected the diameter of myofibers. Myofiber microstructure of longitudinal sections revealed that with the increase of the substitution ratio, the arrangement of myofibers became increasingly sparse. In the 100FSBM group, myofibers even exhibited a breakthrough phenomenon. Dietary glycine supplementation significantly reduced the gap between myofibers, resulting in a tighter arrangement of myofibers in the 75FSBM group (Figure 2). Transmission electron microscopy of longitudinal sections revealed that the Z-line, bright band, and dark band of the sarcomere were clearly visible in the control group. FSBM substitution significantly reduced the length of sarcomeres, whereas dietary glycine supplementation positively promoted an effect on the length of sarcomeres (Figure 3). RT-qPCR results revealed that glycine significantly decreased the mRNA expression of smyhc1 and smyhc2 under 50% FSBM substitution (p < 0.05). However, glycine had no significant effect on the mRNA expression of smyhc1 and smyhc2 under 75% and 100% FSBM substitution (p > 0.05, Figure 4).

3.4. Dietary Glycine Supplementation Relieved Oxidative Stress Induced by FSBM Substitution

In the hepatopancreas, GSH contents were not affected by feed (p > 0.05, Figure 5A). In contrast, compared with the 50FSBM and 75FSBM groups, glycine supplementation significantly decreased the activity of SOD (p < 0.05, Figure 5B). In the muscle, glycine alleviated the increase in SOD and T-AOC activities induced by the 100FSBM group (p < 0.05, Figure 5C,F). In addition, dietary glycine supplementation significantly decreased the contents of GSH induced by 75FSBM and 100FSBM groups (p < 0.05, Figure 5E). However, glycine supplementation had no significant effect on the contents of MDA in the muscle (p > 0.05, Figure 5D).

3.5. Dietary Glycine Supplementation Enhanced the Immunity of Muscle in Shrimp

In addition to enhancing antioxidant capacity, glycine also regulates the immune response. In the present study, compared with the control group, 100% FSBM substitution significantly increased the mRNA expression of lgbp and hsp70 (p < 0.05, Figure 6A,C), whereas glycine supplementation relieved the increase in the control group. Glycine attenuated high-FSBM-induced toll mRNA expression (Figure 6D). The 75FSBM and 100FSBM groups significantly activated the mRNA expression of traf6 (p < 0.05, Figure 6B). After adding glycine, the expression of traf6 decreased in the 75FSBMG group, and there was no significant difference between the 75FSBMG group and the control group (p > 0.05, Figure 6B). However, although glycine also significantly decreased the mRNA expression of traf6 in group 100FSBMG compared with that in the 100FSBM group, it was still significantly higher than that in the control group. Glycine supplementation had no significant effect on the mRNA expression of lysozyme or imd (p > 0.05, Figure 6E,F).