Grouper fish bones were sourced from a processing factory in the Yong’an District of Kaohsiung, Taiwan. The bones were first defrosted and then boiled for 10 min in water to facilitate the separation of residual flesh and connective tissues. After boiling, debridement was carried out to ensure thorough removal of organic matter, after which the bones were weighed. An ultrafine bubble cleansing was then done for 5 min (Chen et al., 2020) for deodorization and sterilization.

The decontaminated bones were then dried with hot air, weighed again, and ground into a fine powder. This powder was preserved at-20°C for later use. The bone powder was later macerated for 12 h in a 0.25% citric acid solution with a solid-to-liquid ratio of 1:4 and continuous agitation. This was done to demineralize the bone matrix and prepare it for extraction. The material was extracted for 20 h at 65°C with a 1:1 ratio of distilled water under constant stirring to ensure optimal diffusion. After extraction, the resultant mixture was filtered, and a gelatinous sample was obtained.

Next, 0.5% FoodPro®30 L enzyme (International Flavors & Fragrances Inc., NY., United States) was added to the sample, which was then extracted again for 6 h at 50°C in distilled water. The enzymatic treatment was halted by thermal deactivation, in which the extract was held at 100°C for 30 min to ensure complete enzyme inactivation. Finally, centrifugation was done for 15 min at 10,000 × g at 4°C, and the clarified GBH supernatant was obtained.

The amino acid components of the GBH were separated using high-performance liquid chromatography (HPLC) in a C18 column (Hypersil GOLD C18, 100 × 2.1-mm inner diameter, particle size of 1.9 μm). The mobile phase consisted of (A) sodium dihydrogen phosphate and (B) acetonitrile and methanol (1:1) with a flow rate of 0.5 mL/min. Gradient analysis was conducted using the method reported by previous (Innocente et al., 2007).

A comprehensive comparison of literature and databases was done to predict the possible types of each mass-to-charge ratio (m/z) and identify potential peptide sequences. The results of the mass spectrometry can be confirmed through several resources. The amino acid mass table provided by Biosyn was used for initial comparisons. The NIST Online Chemistry Molecular Weight Search Engine was used for identifying molecular weights. The proteomics database provided by NCBI was used to match and confirm the approximate peptide sequences.

Male Institute of Cancer Research (ICR) mice (6 weeks old; average weight 27 g) were purchased from BioLASCO (Taipei, Taiwan). The mice were given Laboratory Rodent Diet 5,001 and distilled water ad libitum. They were housed at room temperature (24 ± 2°C) under humidity-controlled conditions (60 ± 5%) and a regular 12-h light/dark cycle. The animal protocol (A11202) was reviewed and approved by the Institutional Animal Care and Use Committee (IACUC) of Fu Jen Catholic University, New Taipei City, Taiwan.

The recommended daily intake of GBH is approximately 200–1,000 mg per day per mouse based on the previous method (Nirmal et al., 2022). Thus, the mouse dose per kilogram of body weight as follows: 500/60 × 12.3 = 103 mg/kg (GBH-1X). Accordingly, the GBH-2X dose was 205 mg/kg, and the GBH-5X dose was 513 mg/kg.

After day 26 of GBH supplementation, blood samples were collected at 15 min after the mice performed swimming exercises. Blood was centrifuged for 10 min (1,500 × g) at 4°C, and sera were collected for analysis. Lactate, ammonia, CK, glucose, BUN, and LDH levels were determined using an autoanalyzer (Hitachi 7,600; Hitachi Co. Ltd., Japan).

On day 27 of the supplementation period, 30 min after GBH supplementation, the forelimb grip strength of the mice was tested according to the method described by Chen et al. (2016). A low-force testing system (PicoScope 2000, Pico Technology Limited, Cambridgeshire, United Kingdom) was used to measure the forelimb absolute grip strength. Animals were acclimated, then gently held by the tail to allow only their forelimbs to grasp the bar. Once a firm grip was established, animals were pulled back until they released the bar, and peak force was recorded. Each animal underwent 3 trials with 1 min rest intervals, the highest recorded force (in grams) was used as an indicator of the grip strength.

A treadmill exhaustion test was conducted based on a previous study using a mouse treadmill with an incline that started at 10^o (Bio-Cando Incorporation, Taipei, Taiwan). The test was performed with a 10% incline, and the speed and duration were gradually increased as follows: 10 m/min for 5 min, 16 m/min for 10 min, 21 m/min for 10 min, 26 m/min for 10 min, 31 m/min for 10 min, 36 m/min for 10 min, and 41 m/min for 10 min until exhaustion. The running time for each mouse was recorded.

At the end of the study on day 28, the fat-free mass (FFM) and fat mass of the mice were detected using time-domain nuclear magnetic resonance (TD-NMR) with an LF50 body composition analyzer (Minispec; Bruker, Germany). Briefly, animals were restrained in a 50-mm-diameter plastic cylinder and scanned for 2 min. The TD-NMR device measures radiofrequency signals from hydrogen nuclei in tissues and uses differences in relaxation times to estimate body composition. Glycogen content was analyzed using a chemical method based on literature protocols, with commercial glycogen (Sigma, United States) as the calibration standard. Tissue samples were homogenized with five volumes (w/v) of buffer using a Bullet Blender (Next Advance, MA, United States). Homogenates were centrifuged at 12,000 × g for 15 min at 4°C, and the supernatant was collected for direct glycogen quantification (Chen et al., 2020).

On day 28, all mice were sacrificed using 95% CO₂ asphyxiation. Different tissues were collected and fixed in 10% formalin. Hematoxylin and eosin staining was carried out as previously described. Blood was withdrawn through cardiac puncture, and serum was collected through centrifugation. The levels of aspartate aminotransferase (AST), alanine aminotransferase (ALT), LDH, CK, creatinine, total cholesterol (TC), triacylglycerol (TG), high-density lipoprotein (HDL), low-density lipoprotein (LDL), total protein (TP), albumin, and glucose were assessed using an autoanalyzer (Hitachi 7,600; Hitachi Co. Ltd., Japan).

Data were expressed as the mean ± standard deviation (SD). Significant differences between multiple groups were calculated using one-way analysis of variance (ANOVA) and Duncan’s post-hoc test. Values of p < 0.05 were considered significant. The Cochran–Armitage trend test was used to examine the dose-dependence effect.

Table 1 presents the total BCAAs content in GBH. Following enzymatic hydrolysis, the bone material exhibited a high concentration of amino acids, and the contents of the BCAAs valine, isoleucine, and leucine were measured as 644 ± 64, 260 ± 26, and 603 ± 57 μg/mL, respectively. The substantial presence of BCAAs in GBH suggests that it has potential to stimulate protein synthesis in skeletal muscle. Thus, it may exert anabolic effects on protein metabolism, potentially enhance the rate of protein synthesis, and decrease protein degradation in resting human skeletal muscle. Table 2 shows the hydrolysis efficiency of GBH. The results indicate a hydrolysis rate of 29.95 ± 1.08%, and the peptide concentration in the GBH was 117.03 ± 2.94 mg/mL.

Table 3 displays the initial peptide test results of GBH, which indicate the potential presence of dipeptides or tetrapeptides. MALDI-TOF mass spectrometry analysis identified five peaks at retention times of 0.9, 1.3, 8.7, 11, and 11.9 min. It is important to note that these findings are preliminary, and due to the nature of grouper fish bones as a residual food resource, the consistency of the sample source cannot be guaranteed. Therefore, the identification of peptides is based on inference. These peptides included dipeptides Leu-Ala, Glu-Asp., Met-Leu, Met-Ile, Phe-Pro, Trp-Asp., Leu-Val, and Leu-Cys and tetrapeptides Pro-Ser-Met-Ala, Ser-Val-Pro-Ile, and Ala-Val-Pro-Trp.

The proposed peptide sequences were determined through a comparison of mass spectrometry data with the amino acid mass table, molecular weight databases, and proteomics databases. Additional validation is needed to confirm these peptide sequences. Leu: leucine, Ala: alanine, Glu: glutamic acid, Asp: aspartic acid, Met: methionine, Phe: phenylalanine, Pro: proline, Trp: tryptophan, Ser: serine, Val: valine, Cys: cysteine, Am: amide, His: histidine, Ile: isoleucine.

Effect of supplementation on tissue weight, body weight, and food and water intake.

Figure 1 illustrates the growth curves of mice supplemented with various doses of GBH (1X, 2X, and 5X) compared to a vehicle control group over 4 weeks. All groups exhibited a steady increase in body weight throughout the experimental period. However, no statistically significant differences in body weight were observed between the GBH-supplemented groups and the vehicle control group at any time point.

Table 4 displays the data on body weight, tissue weights, and food and water consumption. There were no significant differences in any weights or consumption across groups, indicating that GBH supplementation at the tested doses did not significantly affect the overall body weight gain and diet of mice. Table 4 also shows the measured epididymal fat pad (EFP) weights. The weights of the vehicle, GBH-1X, GBH-2X, and GBH-5X groups were 0.50 ± 0.05 g, 0.39 ± 0.09 g, 0.36 ± 0.06 g, and 0.40 ± 0.13 g, respectively. Trend analysis revealed a significant dose-dependent decrease in EFP weight (p = 0.0297). The EFP weights were significantly lower in the GBH-1X (21.88%, p = 0.0287), GBH-2X (29.26%, p = 0.0047), and GBH-5X groups (19.89%, p = 0.0449) compared with the vehicle group.

The weights of muscle tissue from the vehicle, GBH-1X, GBH-2X, and GBH-5X groups were 0.36 ± 0.05, 0.40 ± 0.07, 0.46 ± 0.04, and 0.42 ± 0.03 g, respectively. The muscle weight of the GBH-2X and GBH-5X groups were 1.27 times (p = 0.0012) and 1.16 times (p = 0.0369) higher than that of the vehicle group. The EFP and muscle relative weights of GBH-treated groups were significantly lower than that of the vehicle group (p = 0.0197 and p = 0.0176, respectively).

There were no significant differences in the liver, kidney, and lung weights among groups. Additionally, the relative weights of the liver, kidneys, and lungs were not significantly different. Heart weight was significantly lower in the GBH-2X (p = 0.0107) and GBH-5X (p = 0.0143) groups compared to the vehicle group, but the relative weights of muscle were not significantly different [𝐹 (2, 27) = 2.63, 𝑝 = 0.0733].

Figure 2 presents the effects of GBH supplementation on body composition. There was a significant increase in FFM in the GBH-2X group (1.06 times, p = 0.0217) compared to the vehicle group (Figure 2A). Conversely, fat mass was significantly reduced in all GBH-supplemented groups compared to the vehicle group [GBH-1X: 19.70% (p < 0.0001), GBH-2X: 20.14% (p < 0.0001), GBH-5X: 27.77% (p < 0.0001; Figure 2B)].

The vehicle, GBH-1X, GBH-2X, and GBH-5X groups’ forelimb grip strengths were 126.0 ± 16.6, 142.0 ± 11.4, 149.6 ± 6.5, and 154.1 ± 16.2 g, respectively (Figure 3A). The GBH-2X and GBH-5X groups exhibited 1.19 times (p = 0.0146) and 1.22 times (p = 0.0133) higher grip strength than the vehicle group, respectively. Additionally, a significant dose-dependent effect on relative forelimb grip strength was observed (p < 0.0001).

As illustrated in Figure 3B, the mean exhaustive treadmill test time of the GBH-2X group was 39.2 ± 5.5 min, which was 1.19 times longer than that of the vehicle group (p = 0.0402). The time of the GBH-5X group was significantly longer (1.25 times) at 41.1 ± 4.2 min compared to the vehicle group (p = 0.0091). Trend analysis revealed a statistically significant dose-dependent increase in the treadmill test time (p < 0.0001).

A 15-min swimming test was used to evaluate fatigue-related indicators (Figure 4A). Compared to the vehicle group (11.06 ± 1.58 mmol/L), serum lactate levels were significantly lower in the GBH-2X (7.79 ± 1.58 mmol/L) and GBH-5X (7.63 ± 1.39 mmol/L) groups, with reductions of 29.56% (p = 0.0007) and 30.96% (p = 0.0005), respectively. Furthermore, a significant dose-dependent decrease (p < 0.0001) in blood lactate levels was observed following the 15-min swimming test.

The serum ammonia levels in the GBH-1X (210 ± 11 μmol/L), GBH-2X (181 ± 18 μmol/L), and GBH-5X (171 ± 39 μmol/L) groups were 12.05% (p = 0.0379), 23.99% (p = 0.0002), and 28.03% (p < 0.0001) lower than that in the vehicle group (238 ± 20 μmol/L), respectively. A significant dose-dependent effect on the serum ammonia levels was observed (p < 0.0500; Figure 4B). In the GBH-1X (506 ± 370 U/L), GBH-2X (449 ± 217 U/L), and GBH-5X (274 ± 103) groups, the serum CK levels were significantly lower (64.23%, p < 0.0001; 68.25%, p < 0.0001; and 80.66%, p < 0.0001) compared to the vehicle group (1,414 ± 220 U/L). A significant dose-dependent effect on the CK level was also observed (p < 0.0500; Figure 4C). In contrast, the glucose levels in the GBH group did not show a significant difference from the vehicle group (p = 0.9332; Figure 4D).

As shown in Figure 5A, the liver glycogen content of the GBH-1X, GBH-2X, and GBH-5X groups exhibited a statistically significant elevation relative to the vehicle group. The fold-changes were 1.37 (p = 0.0482) for GBH-1X, 1.42 (p = 0.0582) for GBH-2X, and 1.80 (p = 0.0019) for GBH-5X, corresponding to respective values of 17.5 ± 0.9 mg/g liver, 18.2 ± 5.3 mg/g liver, and 23.0 ± 3.5 mg/g liver, respectively. Similarly, muscle glycogen levels were elevated in the GBH-1X, GBH-2X, and GBH-5X groups compared to the vehicle group, as shown in Figure 5B. The fold-changes were 1.74 (p = 0.0193) for GBH-1X, 1.71 (p = 0.0082) for GBH-2X, and 2.25 (p < 0.0001) for GBH-5X, with respective values of 1.5 ± 0.4 mg/g liver, 1.5 ± 0.7 mg/g liver, and 1.9 ± 0.8 mg/g liver, respectively. Notably, a clear and significant dose-dependent effect on both liver and muscle glycogen content was observed (p < 0.0001).

Representative histological sections from each excised tissue are presented in Figure 6, which shows the morphological and cellular characteristics that appeared following GBH supplementation. No significant differences were observed between GBH-supplemented groups and the vehicle control group in any of the examined tissues, indicating that 4 weeks of GBH supplementation did not induce any adverse effects on the evaluated organs.

Table 5 shows that GBH supplementation did not induce significant changes in liver function markers (AST, ALT, and LDH) compared to the vehicle group (p > 0.05). However, a decreasing trend was observed in CK levels with increasing GBH dosage, suggesting a potential protective effect of GBH against muscle damage (p = 0.2170). GBH supplementation significantly reduced creatinine levels in a dose-dependent manner (p < 0.0001), indicating improved kidney function. No significant changes were observed in other parameters, including TC, TG, HDL, TP, albumin, and glucose levels (p > 0.05). These results suggest that GBH supplementation may have beneficial effects on kidney function and potentially muscle health and lipid metabolism, warranting further investigation.