Preparation, Structural Characterization, and Stability of Low-Molecular-Weight Collagen Peptides–Calcium Chelate Derived from Tuna Bones

This study was conducted to prepare calcium chelate of low-molecular-weight tuna bone collagen peptides (TBCPLMW) with a high chelation rate and to identify its structural characteristics and stability. The optimum conditions for calcium chelation of TBCPLMW (TBCPLMW-Ca) were determined through single-factor experiments and response surface methodology, and the calcium-chelating capacity reached over 90% under the optimal conditions. The amino acid compositions implied that Asp and Glu played important roles in the formation of TBCPLMW-Ca. Structural characterizations determined via spectroscopic analyses revealed that functional groups such as -COO−, N-H, C=O, and C-O were involved in forming TBCPLMW-Ca. The particle size distributions and scanning electron microscopy results revealed that folding and aggregation of peptides were found in the chelate. Stability studies showed that TBCPLMW-Ca was relatively stable under thermal processing and more pronounced changes have been observed in simulated gastric digestion, presumably the acidic environment was the main factor causing the dissociation of the TBCPLMW-Ca. The results of this study provide a scientific basis for the preparation of a novel calcium supplement and is beneficial for comprehensive utilization of tuna bones.


Introduction
Calcium is a vital nutrient for human health, accounting for about 1.5-2% of human body weight [1].Calcium deficiency may provoke a series of diseases, such as osteopenia, osteoporosis, high blood pressure, and kidney stones [2].Currently, there are a variety of commercial calcium supplements, including inorganic calcium, organic calcium [3,4].Nevertheless, the poor water solubility and relatively low absorption rate of calcium lead to its low bioavailability in vivo [5][6][7].Amino acid chelates have been reported to have a higher absorption rate [8], but the use of high-purity amino acids is quite expensive and may produce uncontrollable colour generation and oxidation reactions [9,10].Studies have shown that food-derived peptides chelated with minerals have high stability, absorption, and bioavailability [9,11].Calcium combined with peptides could form a complex with the advantages of fast absorption, less saturation, and lower energy consumption compared to amino acid calcium [12].Therefore, peptides-calcium chelates have promising potential as functional food additives [13].
In recent years, calcium chelates of food-derived peptides have attracted significant attentions owing to their high stability, absorption rate, and bioavailability [5].Various calcium-binding peptides have been prepared from aquatic organism.Wu et al. developed novel octopus scraps peptides with calcium-chelating activity [14].Cui et al. identified the calcium-binding peptides from sea cucumber eggs and explored their calcium-binding modes [15].Some laboratories have conducted preparation technology studies as well as structural and functional studies on calcium chelation with fish collagen peptides from Pacific cod bone [16], European eel bone [17], and tilapia bone [18].Furthermore, it has also been found that lower-molecular-weight peptides have better chelating effects [19,20].Tuna is a highly commercial marine fish that contributes significantly to the development of world fisheries.The processing of canned tuna generates plenty of by-products, including bones, viscera, red meat, and so on, which are usually processed into fishmeal or discarded as waste [21].In particular, tuna bones account for a large proportion of the by-products, and they are rich in collagen and calcium, but their utilization is extremely inadequate [22,23].Previous studies have demonstrated that collagen peptides have antioxidant, osteogenic, immunomodulatory, and anti-inflammatory activities [5,7,18].Therefore, it is critical to utilize the abundant tuna bone resources to produce valuable collagen peptides-related products.
In this research, peptides-calcium chelates were prepared from low-molecular-weight tuna bone collagen peptides (TBCP LMW ).Chelating conditions of TBCP LMW and calcium ions were optimized.Structural characterizations and morphological analysis were also investigated.The stability of the peptides-calcium chelate was assayed under different conditions and through simulated gastrointestinal digestion in vitro.The findings of this study would provide a scientific basis for the application of tuna bone collagen peptidescalcium chelate (TBCP LMW -Ca) and would improve the utilization of tuna bones.

Preparation of TBCP LMW
TBCP LMW was extracted according to the methods developed by our laboratory previously.Details were as follows: Tuna bones were cleaned to remove any remaining flesh and sinew under water, and then they were cut into 2-3 cm pieces with scissors.The cleaned bones were mixed with distilled water (1:10, w/v) and treated with pepsin (2%) at 55 • C for 3 h (pH 3.0).Pepsin was subsequently inactivated at 100 • C for 15 min.The enzymatically digested bones were steamed under 121 • C, 0.12 MPa for 30 min to remove the fat and reduce the hardness and toughness of tuna bones.After filtering the residue, the cleaned bones were placed in 2% edible sodium bicarbonate solution for 120 min and then dried and smashed into powders.The tuna bone powders were added to distilled water at a ratio of 1:10 (w/v), enzymatically digested by adding animal protease (2%, pH 7.0) at 50 • C for 8 h, and then inactivated at 100 • C for 15 min.The supernatant was collected after centrifugation at 11,800× g (TG16-WS, Cenen, Changsha, China) for 15 min (4 • C) and passed through the spiral-wound membranes with cutoff of 1000 Da and 250 Da, respectively, by using multi-functional rolled membrane pilot equipment (RNF-0460, Starmem, Xiamen, China).Then, the fractions in 250-1000 Da solution were collected and lyophilized.

Preparation of TBCP LMW -Ca
Single-factor experiments were conducted, taking the calcium-chelating capacity as a screening index.The prepared TBCP LMW was dissolved in 10 mL ultrapure water, and then it was mixed with different mass ratios of CaCl 2 (1:2, 1:1, 2:1, 3:1, 4:1) at 30 • C, 50 • C, 70 • C, and 90 • C for 1 h, 2 h, 4 h, and 6 h.pH was adjusted to 5.0, 6.0, 7.0, 8.0, and 9.0.After chelation, anhydrous ethanol (9-fold volume) was added to separate the chelate.The mixture was then centrifuged at 7552× g (TG16-WS, Cenen, Changsha, China) for 10 min, and the precipitate was collected and freeze-dried to obtain TBCP LMW -Ca.A three-level, four-factor Box-Behnken Design was carried out to determine the optimal conditions based on the one-way test using four independent variables: peptide/calcium mass ratio (A), pH (B), chelation time (C), and chelation temperature (D) (Table 1).The total amount of calcium in the chelating solution and the calcium content in the supernatant after centrifugation was both determined via the EDTA titration method [24].The calcium-chelating capacity was calculated as follows: Chelation rate (%) = Total calcium ions content/mg − Free calcium ion content/mg Total calcium ions content/mg × 100

Amino Acid Composition Analysis
The TBCP LMW and TBCP LMW -Ca were transferred into the hydrolysis tube, respectively, and 15 mL of hydrochloric acid solution (6 mol/L) was added to mix the samples thoroughly, followed by 3-4 drops of phenol for hydrolysis.The tubes were pumped close to 0 Pa, filled with nitrogen 3 times/min to remove the air and then sealed while full of nitrogen.The sealed hydrolysis tubes were transferred to an electric blast thermostat at 110 • C, 22 h.After being filtered through a 0.22 µm filter, the solutions were analysed using an automatic amino acid analyser (L8900, Hitachi, Tokyo, Japan).

Ultraviolet Spectroscopic Determination
TBCP LMW and TBCP LMW -Ca were prepared into an aqueous solution at a mass concentration of 1 mg/mL.Change in UV absorption wavelength was measured by UV spectrophotometer (Cary60, Agilent, Santa Clara, CA, USA) at the scanning wavelength of 190-400 nm.

Fourier-Transform Infrared (FTIR) Spectroscopy Analysis
The experiments were performed in dry air at (25 ± 1) • C. A 1 mg sample was firstly mixed with KBr and then pressed into a thin transparent sheet.FTIR spectra of TBCP LMW and TBCP LMW -Ca were observed using an FTIR spectrometer (TensorII, Bruker, Saarbrucken, Germany) at 400-4000 cm −1 , and 64 scans of each spectrum were collected.

Particle Size Distribution and Zeta Potential Analysis
The samples were dissolved in ultrapure water and configured to 1 mg/mL.The size distribution and ζ-potential were figured by Zetasizer Nano instrument (ZS90, Malvern Instruments Ltd., Malvern, UK) at 25 • C.

Scanning Electron Microscopy
The microscopic morphology of TBCP LMW and TBCP LMW -Ca was observed using a scanning electron microscope (Sigma 300, ZEISS, Oberkochen, Germany).Sample powders were placed on the conductive adhesive holder and sprayed with gold at 10 mA, and the samples were observed at 8000× magnification under an accelerating voltage of 2.0 kV.

Acid-Base Stability
The lyophilized powder of TBCP LMW -Ca was dissolved in deionized water to 10 mg/mL in each tube.pH was 3, 4, 5, 6, 7, 8, and 9, respectively.The reaction was carried out in a water bath at 37 • C for 1 h.After the reaction, calcium ion content was analysed via the colourimetric method using o-phenolphthalein [14].The acid-base stability of TBCP LMW -Ca was calculated as calcium retention rate of the chelate after the reaction.

Thermal Stability
The lyophilised powder of TBCP LMW -Ca was dissolved in deionised water (pH 7).The final concentration of each tube was 10 mg/mL.The reaction was conducted for 1 h at 50 • C, 60 • C, 70 • C, 80 • C, 90 • C, and 100 • C, respectively.The thermal stability of TBCP LMW -Ca was calculated as calcium retention rate of the chelate after the reaction.

Simulated Digestion In Vitro
A standard static in vitro digestion model (INFOGEST) [25] was applied with minor modifications to assess the stability of TBCP LMW -Ca in simulated gastrointestinal digestion.Stability was calculated as the retention of calcium in TBCP LMW -Ca after digestion.In the gastric phase, TBCP LMW -Ca was dissolved using the simulated gastric fluid electrolyte in a 1:1 ratio.Porcine pepsin was added to achieve an activity of 2000 U/mL.At the same time, pre-prepared gastric lipase solution was added to achieve 60 U/mL gastric lipase activity in the mixture.After regulating the pH to 3, the TBCP LMW -Ca concentration was adjusted to 10 mg/mL.In the intestinal stages, TBCP LMW -Ca was dissolved in a 1:1 ratio using a simulated intestinal fluid electrolyte and added to the hepatic bile water compound for 30 min.Meanwhile, trypsin, pancreatic rennet, pancreatic α-amylase and pancreatic lipase were added to reach 100, 25, 200, and 2000 U/mL, respectively.pH was adjusted to 7, and the reaction was timed from this point onwards.In the gastrointestinal phase, all previous steps were repeated, with a simulated gastric reaction for 90 min, followed immediately by adjustment of pH to add the enzyme required for the intestinal phase.Each stage of the reaction was carried out at 37 • C in a constant temperature water bath shaker.The enzyme was then inactivated at 100 • C for 15 min.The content of calcium ions in the solution was measured every 30 min using the colourimetric method with o-phenolphthalein.Stability of TBCP LMW -Ca in simulated gastrointestinal digestion was calculated and expressed as calcium retention rate after digestion.

Statistical Analysis
All measurements were performed thrice in parallel, and all data were presented as mean ± standard deviation.SPSS 26 was used for data analysis.Statistical significance was determined via Duncan's multiple range test of one-way ANOVA analysis.p < 0.05 was considered as statistically significant.

Single Factor Experiments
The calcium chelates of collagen peptides extracted from various sources have been reported as potential calcium supplements, such as cod [26], tilapia [27], sheep bone, and bovine bone [28,29], and the processing conditions could affect the calcium-chelating capacity of the peptides.
In the study, the effects of different factors on calcium chelation rate were investigated.As depicted in Figure 1A, when the peptide/calcium mass ratio increased gradually, the capacity to chelate calcium increased (p < 0.05), and there was no significant difference in calcium-chelating ability when the mass ratio exceeded 3:1.According to Figure 1B, when the pH value varied between 5 and 8, the calcium-chelating capacity firstly increased from 5 to 6 (p < 0.05), peaked at pH 6 with a chelating capacity of 90.47%, and decreased gradually when the pH exceeded 6 (p < 0.05).It implied that pH value effected the calcium chelation rate significantly, consistent with the conclusion of Luo et al. [5].The change might be owing to the fact that as the pH value increased, OH − competed with the electron donor group for Ca 2+ , preferentially binding to produce calcium hydroxide precipitates, which was not conducive to chelate.As the pH value decreased, the interaction between Ca 2+ and peptides would be destroyed by the large amount of H + [5].As shown in Figure 1C, the chelation reaction proceeded rapidly in a short period of time.The calciumchelating capacity was higher when the reaction time was 1 h or 2 h than that of 4 h and 6 h (p < 0.05).It can be speculated that the prolonged reaction time may have led to some decomposition of TBCP LMW -Ca and thus reduced the calcium-chelating capacity.According to Figure 1D, when the temperature arose from 30 • C to 70 • C, the calciumchelating capacity increased, and it reached the maximum at 70 • C. The calcium-chelating capacity declined notably when the temperature rose to 90 • C. Chelation reaction is a dynamic equilibrium process.An appropriate temperature will accelerate the molecular motion and improve the chelation rate, while an excessive temperature would change the conformation of peptides and hinder chelation [29,30].
°C.The calcium-chelating capacity declined notably when the temperature rose to 90 °C.Chelation reaction is a dynamic equilibrium process.An appropriate temperature will accelerate the molecular motion and improve the chelation rate, while an excessive temperature would change the conformation of peptides and hinder chelation [29,30].

Response Surface Optimization
Experiments were designed with the peptide/calcium mass ratio (A), pH (B), chelation time (C), and chelation temperature (D) as response variables, with calciumchelating capacity as the response value, with three levels for each variable and equidistance between each level.A total of 29 sets of experiments were conducted, including five sets of centre point replicates, and the results are displayed in Table 2.

Response Surface Optimization
Experiments were designed with the peptide/calcium mass ratio (A), pH (B), chelation time (C), and chelation temperature (D) as response variables, with calcium-chelating capacity as the response value, with three levels for each variable and equidistance between each level.A total of 29 sets of experiments were conducted, including five sets of centre point replicates, and the results are displayed in Table 2.The results of the RSM experiments were analysed.Multiple regression was fitted to give the following quadratic multinomial regression model equation: The model was subjected to an ANOVA, and the results are depicted in Table 3. p-value of the regression equation model was lower than 0.01, indicating a good fit of the equation.The lack-of-fit value of 0.1110 was not significant relative to the pure error, which indicated that the model was a reasonable choice.The correlation coefficient (R 2 = 0.9908) was relatively reliable.Therefore, this model could be used to analyse and predict the chelation of TBCP LMW with calcium chloride.Among the variables, peptide/calcium mass ratio, pH, and temperature all had extremely significant effects on the chelation rate (p < 0.01), while the effect of chelation time was not significant (p > 0.05).The squared terms of all factors were extremely significant (p < 0.01).According to the model, the following optimum chelating conditions were recommended: peptides/calcium mass ratio 3.29:1, pH 6.14, time 2.04 h, temperature 68.69 • C, with a theoretical chelation rate of 94.56%.For practical considerations, the predicted process was modified: peptide/calcium mass ratio 3.3:1, pH 6.1, time 2 h, and temperature 69 • C. The maximum actual chelation rate measured under these conditions was 94.27%.The chelation rate of this optimum model was slightly higher than the findings of Wu et al. [31].The use of TBCP LMW to prepare calcium chelate under this optimum condition was shown to be advantageous and worthy of further study.

Amino Acid Composition
Amino acid compositions of TBCP LMW and TBCP LMW -Ca were presented in Figure 2A.They were abundant in Glu, Asp, Gly, Hyp, Pro, etc., which confirmed the characteristics of collagen peptides, and frequently present in the mineral-binding peptides [5,32].Compared to TBCP LMW , TBCP LMW -Ca showed a significant increase in the contents of Asp and Glu.According to previous studies, acidic amino acids have been recognized as essential amino acids, and they may facilitate the calcium-chelating capacity of proteins and peptides, owing to the free carboxyl groups [24,33].It was in line with the research of Zhang et al. that Asp was one of the important components involved in the calcium chelation of the decapeptide from Pacific cod bone hydrolysate, and Glu formed peptide calcium chelates by binding to calcium ions through multiple carboxylic acid groups [16].All the results indicated that Asp and Glu played important roles in the formation of TBCP LMW -Ca.
According to the model, the following optimum chelating conditions were recommended: peptides/calcium mass ratio 3.29:1, pH 6.14, time 2.04 h, temperature 68.69 °C, with a theoretical chelation rate of 94.56%.For practical considerations, the predicted process was modified: peptide/calcium mass ratio 3.3:1, pH 6.1, time 2 h, and temperature 69 °C.The maximum actual chelation rate measured under these conditions was 94.27%.The chelation rate of this optimum model was slightly higher than the findings of Wu et al. [31].The use of TBCPLMW to prepare calcium chelate under this optimum condition was shown to be advantageous and worthy of further study.

Amino Acid Composition
Amino acid compositions of TBCPLMW and TBCPLMW-Ca were presented in Figure 2A.They were abundant in Glu, Asp, Gly, Hyp, Pro, etc., which confirmed the characteristics of collagen peptides, and frequently present in the mineral-binding peptides [5,32].Compared to TBCPLMW, TBCPLMW-Ca showed a significant increase in the contents of Asp and Glu.According to previous studies, acidic amino acids have been recognized as essential amino acids, and they may facilitate the calcium-chelating capacity of proteins and peptides, owing to the free carboxyl groups [24,33].It was in line with the research of Zhang et al. that Asp was one of the important components involved in the calcium chelation of the decapeptide from Pacific cod bone hydrolysate, and Glu formed peptide calcium chelates by binding to calcium ions through multiple carboxylic acid groups [16].All the results indicated that Asp and Glu played important roles in the formation of TBCPLMW-Ca.

Molecular Weight Distribution
Molecular weight distribution of TBCP LMW and TBCP LMW -Ca were presented in Figure 2B, and the chromatograms are displayed in Figure S1.TBCP LMW was mainly concentrated below 1000 Da, with 250-1000 Da accounting for 82.75%.It has been reported that fish collagen peptides between 180 and 2000 Da have better calcium-chelating ability [19], and peptides with smaller molecular weights are more favourable for chelating calcium ions [34].Therefore, in this study, low-molecular-weight peptides were selected for the preparation of peptide calcium chelate.
Although a certain calcium dissociation of TBCP LMW -Ca existed in the mobile phase due to the acidic environment, the molecular weight distribution of TBCP LMW -Ca complex differed from TBCP LMW -Ca.As depicted in Figure 2B, the percentage of 1000-1500 Da was 32.9% after chelation, while that of TBCP LMW was 5.1%.The main reason may be the formation of new chelation bonds, which increased the peptide crosslinks.Similar changes Foods 2023, 12, 3403 9 of 15 in molecular weight were reported during the preparation of calcium chelates from bovine collagen peptides [35].

Fluorescence Spectroscopy
Structural changes between amino acid groups and mineral ions could be judged by the wavelength and intensity changes in fluorescence spectroscopy [36].The fluorescence results in Figure 3A showed that TBCP LMW -Ca showed a red shift and an increase in endogenous fluorescence intensity at the excitation wavelength of 295 nm.This enhancement may be due to the complex interaction between the chromophore and calcium leading to a change in excited state energy, which in turn affected the intensity.This is consistent with the results of previous studies [24,37], which showed that the chelation of calcium by TBCP LMW led to structural changes in amino acids and peptides, resulting in folding and aggregation, and indicated the successful binding of TBCP LMW to calcium ions.
for the preparation of peptide calcium chelate.
Although a certain calcium dissociation of TBCPLMW-Ca existed in the mobile phase due to the acidic environment, the molecular weight distribution of TBCPLMW-Ca complex differed from TBCPLMW-Ca.As depicted in Figure 2B, the percentage of 1000-1500 Da was 32.9% after chelation, while that of TBCPLMW was 5.1%.The main reason may be the formation of new chelation bonds, which increased the peptide crosslinks.Similar changes in molecular weight were reported during the preparation of calcium chelates from bovine collagen peptides [35].

Fluorescence Spectroscopy
Structural changes between amino acid groups and mineral ions could be judged by the wavelength and intensity changes in fluorescence spectroscopy [36].The fluorescence results in Figure 3A showed that TBCPLMW-Ca showed a red shift and an increase in endogenous fluorescence intensity at the excitation wavelength of 295 nm.This enhancement may be due to the complex interaction between the chromophore and calcium leading to a change in excited state energy, which in turn affected the intensity.This is consistent with the results of previous studies [24,37], which showed that the chelation of calcium by TBCPLMW led to structural changes in amino acids and peptides, resulting in folding and aggregation, and indicated the successful binding of TBCPLMW to calcium ions.

UV-Vis Absorption Spectroscopy Assay
Changes in intensity and dislocation in the UV absorption spectra can reflect the differences between peptides and calcium chelate [38].According to Figure 3B, the absorption spectra of TBCP LMW and TBCP LMW -Ca had multiple absorption peaks in the range of 190-280 nm.A higher absorption peak of TBCP LMW occurred at 220 nm, which was consistent with the findings that the n→π* leap of C=O in the peptide bond generally occurs around 210 nm [39].After chelating calcium, it can be seen from the chromatogram that the UV absorption spectrum was shifted considerably, and new substances were produced.Moreover, the absorption intensity of the closing peak of TBCP LMW -Ca at 201 nm was reduced, and a colour reduction effect was observed, which may be the result of the chelation of carboxyl and amino groups with calcium ions [40].The results of the spectra demonstrated that TBCP LMW reacted with calcium ions, and TBCP LMW in the near-UV region represented higher absorbance than that of TBCP LMW -Ca, which may be due to the spatial structure changes in TBCP LMW caused by binding with calcium ions.FTIR spectrum is a good reflection of the differences between two substances, and the changes in the characteristic absorption peaks reflect the interaction of the mineral ions with the organic groups of the protein [41].
According to Figure 3C, the peak of TBCP LMW at 3412.34 cm −1 shifted to 3389.59 cm −1 after calcium chelation, corresponding to the stretching vibration of N-H, indicating that N-H contributed to the formation of the chelate.The amide I band of TBCP LMW shifted from 1653.26 cm −1 to 1656.46 cm −1 , relating to the stretching vibration of the carbonyl group in the peptide bonds, which demonstrated that C=O was involved in the formation of TBCP LMW -Ca.In addition, the amide II peak of TBCP LMW moved from 1542.79 cm −1 to 1538.13 cm −1 in TBCP LMW -Ca, representing the bending vibration of N-H, implying that -NH 2 might react with calcium.The absorption wave number 1401.08 cm −1 moved to a higher frequency of 1409 cm −1 after chelation, while the vibrational spectral region of 1430-1370 cm −1 was associated with the stretching vibrations of carboxylate group.It suggested that -COO − may be bound to calcium and converted to -COOCa, which was consistent with previous findings [36].The 1300-600 cm −1 region represents the fingerprint region of the compound.After the chelation occurred, the peak at 1081.36 cm −1 moved to 1043.60 cm −1 , indicating that the C-O bond might participate in the chelation and formed new coordination bonds with calcium [42].Therefore, it can be speculated that -COO − , N-H, C=O, and C-O groups might be involved in the formation of TBCP LMW -Ca.

Circular Dichroism
The protein secondary structure can be analysed by circular dichroism spectra [43].As shown in Figure 3D, the β-sheet in TBCP LMW -Ca increased from 15.4% to 21.4%, which led to the conclusion that calcium ions could promote β-sheet production [24,34].The binding of calcium to TBCP LMW decreased both α-helix and random folding.It is hypothesized that the binding of calcium may increase the exposure of hydrophobic groups and lead to the decay of intramolecular hydrogen bonds, thus reducing the α-helix as well as increasing the β-sheet.The reduction in random folding suggested that TBCP formed a tighter secondary structure upon binding to calcium ions, similarly to that of previous research by Zhang et al. [16] and Yang et al. [44].

Zeta Potential and Particle Size Distribution
The surface charge of protein particles was assayed through Zeta potential measurement [45].As depicted in Figure 4A, the value of TBCP LMW -Ca was significantly lower compared to TBCP LMW , dropping from 17.67 mv to 4.66 mv.It suggested that a transfer of electrons between TBCP LMW occurred when they reacted with calcium.It could be the negative charge of TBCP LMW was neutralized by calcium ions [46].
Particle size distributions of TBCP LMW and TBCP LMW -Ca are displayed in Figure 4B.The mean particle size of TBCP LMW -Ca increased from 285.13 nm to 313.45 nm.An increase in the mean particle size was observed after chelation, presumably due to the mineral ions acting as a salt bridge by shielding the negative charge on the peptide chain and promoting protein aggregation [37].Particle size distributions of TBCPLMW and TBCPLMW-Ca are displayed in Figu The mean particle size of TBCPLMW-Ca increased from 285.13 nm to 313.45 nm.An in in the mean particle size was observed after chelation, presumably due to the miner acting as a salt bridge by shielding the negative charge on the peptide chai promoting protein aggregation [37].

Scanning Electron Microscope
SEM images of TBCPLMW and TBCPLMW-Ca in magnification of 8000 are displa Figure 4C,D.The surface of TBCPLMW was dense and smooth, while that of TBCPL was rough with a large number of pores and granular aggregates.The differences be TBCPLMW and TBCPLMW-Ca lied in the fact that the calcium chelation had destruct dense structure of TBCPLMW.The amino and carboxyl groups of TBCPLMW were com with calcium ions, inducing protein aggregation.The results were consistent w electron micrographs of bovine collagen peptide calcium chelate [28,35,44].

pH and Temperature
The calcium retention rate of TBCPLMW-Ca at different conditions of pH temperatures are displayed in Figure 5A,B.The calcium retention decreased signifi at pH 3 and 4, indicating that high acidic conditions were not conducive to the ret of calcium in TBCPLMW-Ca.It was due to the fact that the abundant H + may compet Ca 2+ and lead to dissociation of TBCPLMW-Ca [6].A similar situation was found in p collagen peptide calcium chelate [31].On the other hand, the calcium retention TBCPLMW-Ca slightly decreased as the temperature changing from 50 °C to 90 °C remained above 78%.It suggested that the chelate has great resistance to heating, s to the peptide-calcium chelate from stickwater and oyster shells [47].

Scanning Electron Microscope
SEM images of TBCP LMW and TBCP LMW -Ca in magnification of 8000 are displayed in Figure 4C,D.The surface of TBCP LMW was dense and smooth, while that of TBCP LMW -Ca was rough with a large number of pores and granular aggregates.The differences between TBCP LMW and TBCP LMW -Ca lied in the fact that the calcium chelation had destructed the dense structure of TBCP LMW .The amino and carboxyl groups of TBCP LMW were combined with calcium ions, inducing protein aggregation.The results were consistent with the electron micrographs of bovine collagen peptide calcium chelate [28,35,44]  The calcium retention rate of TBCP LMW -Ca at different conditions of pH and temperatures are displayed in Figure 5A,B.The calcium retention decreased significantly at pH 3 and 4, indicating that high acidic conditions were not conducive to the retention of calcium in TBCP LMW -Ca.It was due to the fact that the abundant H + may compete with Ca 2+ and lead to dissociation of TBCP LMW -Ca [6].A similar situation was found in porcine collagen peptide calcium chelate [31].On the other hand, the calcium retention rate of TBCP LMW -Ca slightly decreased as the temperature changing from 50 • C to 90 • C, but it remained above 78%.It suggested that the chelate has great resistance to heating, similar to the peptide-calcium chelate from stickwater and oyster shells [47].

In Vitro Digestion Simulation
Dietary nutrients usually pass through the stomach and reach the intestine, where they are digested, absorbed, and utilised.It has been shown that pepsin in the stomach can hydrolyse peptides into amino acids, resulting in adverse changes in the biological activity of the peptides [48].The tolerance of TBCPLMW-Ca to gastrointestinal enzymes and its stability in the digestive environment was evaluated using the INFOGEST static digestion system.
As shown in Figure 5C, the reaction conditions chosen in the stomach alone were pH 3 with the addition of pepsin as well as gastric lipase, and the retention of calcium decreased to approximately 66.40% at 30 min intervals.This indicated that pH had a large effect on retention, while the addition of pepsin and gastric lipase may cause further hydrolysis of the peptide, so gastric environment may be the main stage for the release of calcium ions.In contrast, TBCPLMW-Ca had a higher stability in the intestinal alkaline digestive environment and was largely unaffected by bile as well as other digestive enzymes [31].In the gastrointestinal phase, the calcium retention rate decreased and remained at about 67.28% for the first 90 min, while it increased significantly when converted to the intestinal environment.It implied that the intestinal environment can not

In Vitro Digestion Simulation
Dietary nutrients usually pass through the stomach and reach the intestine, where they are digested, absorbed, and utilised.It has been shown that pepsin in the stomach can hydrolyse peptides into amino acids, resulting in adverse changes in the biological activity of the peptides [48].The tolerance of TBCP LMW -Ca to gastrointestinal enzymes and its stability in the digestive environment was evaluated using the INFOGEST static digestion system.
As shown in Figure 5C, the reaction conditions chosen in the stomach alone were pH 3 with the addition of pepsin as well as gastric lipase, and the retention of calcium decreased to approximately 66.40% at 30 min intervals.This indicated that pH had a large effect on retention, while the addition of pepsin and gastric lipase may cause further hydrolysis of the peptide, so gastric environment may be the main stage for the release of calcium ions.In contrast, TBCP LMW -Ca had a higher stability in the intestinal alkaline digestive environment and was largely unaffected by bile as well as other digestive enzymes [31].In the gastrointestinal phase, the calcium retention rate decreased and remained at about 67.28% for the first 90 min, while it increased significantly when converted to the intestinal environment.It implied that the intestinal environment can not only inhibit the dissociation of calcium ions from TBCP LMW -Ca, but also contribute to the re-chelation of TBCP LMW and calcium [15].
It was clear from the results that the main factor leading to the dissociation of calcium ions from TBCP LMW -Ca was a change in pH, using an in vitro simulated digestion model combined with pH experiments.Pepsin, gastric lipase, pancreatic protease, and bile had little effect on TBCP LMW -Ca, and the weak alkaline environment in the intestine promoted re-chelation of calcium ions and peptides.

Conclusions
In this study, TBCP LMW -Ca were prepared to use low-molecular-weight collagen peptides obtained via enzymatic digestion combined with membrane grading technique from tuna bones, and the preparation conditions of TBCP LMW -Ca were optimized.The analysis of amino acid compositions showed that Asp and Glu may facilitate the chelation with calcium.The results of structural characterization revealed the changes in TBCP LMW structure after calcium chelation.The molecular weight distribution and morphological analysis indicated that calcium ions cross-linked with collagen peptides, and aggregation occurred in TBCP LMW -Ca.The stability analysis investigated that TBCP LMW -Ca was comparatively stable at high temperatures and gastrointestinal digestion conditions.In conclusion, this study demonstrated the feasibility of using tuna bones to prepare a peptides-calcium chelate, which is nutritionally beneficial and has wild application prospects.

Figure 5 .
Figure 5. Stability of TBCPLMW-Ca: (A) at different pH, (B) at different temperatures, and (C) during simulated gastrointestinal digestion in vitro.a,b,c,d,e,f,g: statistical significance.

Figure 5 .
Figure 5. Stability of TBCP LMW -Ca: (A) at different pH, (B) at different temperatures, and (C) during simulated gastrointestinal digestion in vitro.a,b,c,d,e,f,g: statistical significance.

Table 1 .
Response surface factors and levels of peptides calcium chelation process.

Table 2 .
Response surface design and experimental results.

Table 2 .
Response surface design and experimental results.

Table 3 .
Analysis of variance of regression equation parameters.