Production, identification, in silico analysis, and cytoprotection on H2O2-induced HUVECs of novel angiotensin-I-converting enzyme inhibitory peptides from Skipjack tuna roes

Background Exceeding 50% tuna catches are regarded as byproducts in the production of cans. Given the high amount of tuna byproducts and their environmental effects induced by disposal and elimination, the valorization of nutritional ingredients from these by-products receives increasing attention. Objective This study was to identify the angiotensin-I-converting enzyme (ACE) inhibitory (ACEi) peptides from roe hydrolysate of Skipjack tuna (Katsuwonus pelamis) and evaluate their protection functions on H2O2-induced human umbilical vein endothelial cells (HUVECs). Methods Protein hydrolysate of tuna roes with high ACEi activity was prepared using flavourzyme, and ACEi peptides were isolated from the roe hydrolysate using ultrafiltration and chromatography methods and identified by ESI/MS and Procise Protein/Peptide Sequencer for the N-terminal amino acid sequence. The activity and mechanism of action of isolated ACEi peptides were investigated through molecular docking and cellular experiments. Results Four ACEi peptides were identified as WGESF (TRP3), IKSW (TRP6), YSHM (TRP9), and WSPGF (TRP12), respectively. The affinity of WGESF (TRP3), IKSW (TRP6), YSHM (TRP9), and WSPGF (TRP12) with ACE was −8.590, −9.703, −9.325, and −8.036 kcal/mol, respectively. The molecular docking experiment elucidated that the significant ACEi ability of WGESF (TRP3), IKSW (TRP6), YSHM (TRP9), and WSPGF (TRP12) was mostly owed to their tight bond with ACE’s active sites/pockets via hydrophobic interaction, electrostatic force and hydrogen bonding. Additionally, WGESF (TRP3), IKSW (TRP6), YSHM (TRP9), and WSPGF (TRP12) could dramatically elevate the Nitric Oxide (NO) production and bring down endothelin-1 (ET-1) secretion in HUVECs, but also abolish the opposite impact of norepinephrine (0.5 μM) on the production of NO and ET-1. Moreover, WGESF (TRP3), IKSW (TRP6), YSHM (TRP9), and WSPGF (TRP12) could lower the oxidative damage and apoptosis rate of H2O2-induced HUVECs, and the mechanism indicated that they could increase the content of NO and activity of superoxide dismutase (SOD) and glutathione peroxidase (GSH-Px) to decrease the generation of reactive oxygen species (ROS) and malondialdehyde (MDA). Conclusion WGESF (TRP3), IKSW (TRP6), YSHM (TRP9), and WSPGF (TRP12) are beneficial ingredients for healthy products ameliorating hypertension and cardiovascular diseases.

Hypertension is a clinic-familiar disease affecting the arteries of the human body, and uncontrolled hypertension becomes a huge potential risk of cardiovascular diseases (CVDs), atherosclerosis (AS), heart failure, stroke, and kidney diseases (16). In WHO's report, about 1.28 billion people between 30 and 79 years worldwide have the disease, and the group number will increase to 1.56 billion in 2030 if it is not properly controlled. The causes of high blood pressure are intricate and multifaceted among independent individuals, but evidences continue to confirm that family history, lack of exercise, tobacco use or vaping, excess alcohol consumption, certain chronic conditions, and obesity or overweight induced by a dietary imbalance were dominant factors to the increasing CVD populations (16,17). Healthy lifestyle habits can help control hypertension, but severe patients must be treated with drugs (18). The oral drug is the common therapeutic measure to lower high blood pressure, and developing new drugs is the primary task to effectively control and manage hypertension population (19,20). Angiotensinconverting enzyme (ACE) can inactivate the vasodilator bradykinin to up-regulate blood pressure via modifying angiotensin (Ang) I to active Ang II, then inhibition of ACE activity is a crucial approach to mediate systematic hypertension (21). Therefore, the synthetic ACE inhibitory (ACEi) drugs, including captopril (Cap), lisinopril and enalapril, have been used clinically to treat hypertension, endothelial dysfunction, and diabetic nephropathy, but these ACEi drugs show serious side effects and require careful prescription management (20,22). Therefore, the search for safer ACEi drugs from natural resources can provide feasible alternatives to synthetic ACEi drugs for treating hypertension and CVDs.
Tuna and tuna-like species with catches of 8.0 × 10 6 tons/year are one of the four most highly valuable catch groups worldwide, but their byproducts, composed of viscera, bones, heads and skins, occupy about 70% of processed fish in the factory process, which lead to serious financial losses and environmental contamination (34). Developing active ingredients or products using tuna byproducts is a delightful choice to reduce economic damage, protect the ecological environment, and provide quality products to consumers (34). Therefore, some BPs have been generated from different tuna and its byproducts, such as scale (35), muscle (36,37), cardiac arterial bulbs (38)(39)(40), bone/frame (11,41), skins (42,43), milts (2,44), and head and viscera (45,46). In previous research, antioxidant peptides, such as YEA, ICRD, GEC, AEHNH, AEM, QDHKA, and YVM have been isolated from Skipjack tuna (Katsuwonus pelamis) roes and showed significant activity (47,48). To make more efficient use of tuna roes, the objectives of this research were to produce and identify ACEi peptides from protein hydrolysate of Skipjack tuna roes. Moreover, the cytoprotection of prepared ACEi peptides on endothelial cells (ECs) against oxidative damage was systematically discussed.

Determination of ACEi activity
The ACEi activity was measured by employing FAPGG as the substrate with the following modifications reported by Zhao et al. (18). In brief, the initial assay volume consisted of 50 μL of the substrate (3 mM), 50 μL of the ACE enzyme solution containing 1.25 mU of declared enzyme activity, and 50 μL of assay sample. All these solutions were incubated for 30 min at 37°C in a water bath first without mixing and then for an additional 30 min after mixing. ACE activity was stopped by 150 μL of glacial acetic acid. After that, the reaction mixture was separated by HPLC at 228 nm to determine the hippuric acid (HA) content produced due to ACE activity on the substrate. The control reaction mixture contained 50 μL of buffer instead of the assay sample and the control was expected to liberate the maximum amount of HA from the substrate due to uninhibited ACE activity. The percent inhibition of ACE activity was calculated as follows: The IC 50 value is the concentration of peptide inhibiting 50% activity of ACE.

Preparation of roe hydrolysate of skipjack tuna
The preparation of tuna roe hydrolysate was performed according to the previous method (17). The degreasing process of Skipjack tuna roes was performed according to the described method by Wang et al. (48). The defatted tuna roes were separately hydrolyzed by alcalase (55°C, pH 8.5), neutrase (55°C, pH 7.0), flavourzyme (50°C, pH 7.0), papain (55°C, pH 7.0), pepsin (37.5°C, pH 2.0), and trypsin (37.5°C, pH 7.8) with enzyme dose of 2% (w/w) for 90, 120, 150, 180, 210, 240, or 270 min. After the hydrolysis reaction, the hydrolysates were heated in boiling water for 10 min to inactivate the proteases. Each hydrolysate was centrifuged at 10,000 g for 20 min, and the supernatants were freeze-dried and stored at −20°C. The tuna roe hydrolysate produced by flavourzyme showed the highest ACEi activity and was named STRH.

Separation of ACEi peptides from STRH
ACEi peptides were purified from TMPH using the following designed process (Figure 1).
Frontiers in Nutrition 04 frontiersin.org IEC-III (6.0 mL, 45.0 mg/mL) was purified by a Sephadex G-25 column (2.6 × 150 cm) and eluted with ultrapure water under a flow rate of 1.0 mL/min. Each collected eluate (3.0 mL) was monitored at 220 nm and four subfractions (GPC-1 to GPC-4) were isolated from IEC-III and lyophilized.

Molecular docking experiments of TRP3, TRP6, TRP9, and TRP12
The crystal structures of captopril (Cap) and human ACE-lisinopril complex (1O8A.pdb) were gained from the RCSB PDB Protein Data Bank (PDB code: 1UZF). The Chimera software was used to confirm the position and size of the binding pocket by analyzing the interaction between ACE and peptide (TRP3, TRP6, TRP9, or TRP12). Non-standard residues in 1UZF model were deleted, and PDB files were converted into PDBQT files by the Autodock Tools. ACEi peptides (TRP3, TRP6, TRP9, and TRP12) were converted into SMILES format by the PepSMI tool, 3D models were drawn by the Discovery Studio program, and energy was minimized using steepest descent and conjugate gradient techniques. Molecular docking and free energy calculation were performed using the Autodock Vina. The best-ranked docking poses of TRP3, TRP6, TRP9, and TRP12 in ACE were captured on the binding-energy values and scores.

Cell viability, NO, and ET-1 determination
The viability of HUVECs treated by TRP3, TRP6, TRP9, and TRP12 was determined using MTT method. HUVECs were seeded in the 96-well plates, cultured for 24 h, and treated with ACEi peptides (100, 200, and 300 μM) at 37° C for 24 h. MTT (final content of 2 mg/mL) was added into cell culture. After 4 h, DMSO was added into each well and monitored at 490 nm. The cell viability (% control) was calculated.
After treating with ACEi peptides (TRP3, TRP6, TRP9, or TRP12, respectively) for 24 h, NO and ET-1 contents were separately determined by employing human NO and ET-1 assay kit as per manufactures' protocol (18).

-induced HUVECs
The oxidative damage model of HUVECs was established according to the previous method (50). HUVECs were seeded in the 96-well plates, cultured for 24 h, and treated with 300 mM The HUVECs were cultured for 24 h in a 96-well plate. Subsequently, the supernatant was aspirated and 20 μL of GSH (200 μM) and ACEi peptides (100 or 200 μM) were added in the protection groups, respectively. After 24 h, ACEi peptides were removed and H 2 O 2 with the final concentration of 300 μM was added into the damage and protection groups for incubating 24 h.

Measurement of levels of ROS, SOD, GSH-Px, NO, and MDA
The level of ROS was determined using DCFH2-DA assay and expressed as % control (50). The levels of SOD (U/mgprot), GSH-Px (U/ mgprot), NO (μmol/L), and MDA (nmol/mgprot) were measured using assay kits according to manufacturer' protocols.

Morphological observation of HUVECs
Cell treatments with ACEi peptides, GSH, and H 2 O 2 were according to the above method (50). The morphology of HUVECs was observed and photographed using an inverted microscope (Nikon Corporation, Kyoto, Japan). The percentage of apoptotic HUVECs was analyzed using previous methods (17, 50).

Data analysis
All data are expressed as the mean ± standard deviation (SD, n = 3) and analyzed by SPSS 19.0. An ANOVA test with Dunnett or Tukey test was employed to carry out the significant difference analysis (p < 0.05, p < 0.01, or p < 0.001).

Preparation of protein hydrolysate of Skipjack tuna roes
Proteins of Skipjack tuna roes were hydrolyzed by six proteases and the ACEi rates of generated hydrolysates were shown in Figure 2. The kind of proteases and hydrolysis time significantly influenced the ACEi rates of tuna roe hydrolysates. In addition, the ACEi rates increased gradually when the hydrolysis time ranged from 60 min to 180 min, but subsequently decreased after 180 min. Moreover, the tuna roe hydrolysate with the maximum ACEi rate (62.87 ± 1.98%) was produced by flavourzyme for 180 min, and the hydrolysate (STRH) was selected for the preparation of ACEi peptides.
Frontiers in Nutrition 06 frontiersin.org The effects of TRP3, TRP6, TRP9, and TRP12 on the viability of HUVECs at 100-300 μM were shown in Figure 6A. After incubating for 24 h, the cell viability of TRP3 group ranged from 101.92 ± 3.62% to 86.72 ± 4.45%. It was important to note the cell viability of TRP3 and TRP9 groups at 300 μM was 86.72 ± 4.45% and 91.27 ± 4.05%, which was significantly smaller than those of the control and other groups. The results implied that this concentration (300 μM) might have some negative effects on the proliferation of HUVECs. Therefore, 100 and 200 μM were selected as the test concentrations of TRP3, TRP6, TRP9, and TRP12 in the follow-up experiments.
3.5.2. Effects of TRP3, TRP6, TRP9, and TRP12 on No and ET-1 production NO deficiency will give rise to the risks of CVDs, and improving the production of endothelial NO represents a good therapeutic approach for atherosclerosis (17, 24). Compared with the control group, the NO level in HUVECs was significantly increased from 33.68 ± 0.96 μM to 53.71 ± 2.313 μM by Cap treatment, but significantly decreased to 18.59 ± 0.82 μM by NE treatment (p < 0.001) ( Figure 6B)   Frontiers in Nutrition 08 frontiersin.org TRP9, and TRP12 could significantly increase the NO production in HUVECs and offset in part of the decreased content by NE. As a functional factor similar to Ang II, ET-1 can lead to endothelial dysfunction correlated with coronary heart disease and hypertension (18, 22). As depicted in Figure 6C, the ET-1 content in HUVECs was significantly decreased from 55.37 ± 2.53 pg./mL (control group) to 27.32 ± 1.28 pg./mL by Cap (1.0 μM) treatment and increased to 73.57 ± 3.09 pg./mL by NE (0.5 μM) treatment (p < 0.001). Furthermore, the ET-1 content in HUVECs significantly (p < 0.001) decreased by TRP3, TRP6, TRP9 and TRP12 at 100 and 200 μM, and the ET-1 content in TRP3, TRP6, TRP9, and TRP12 groups reduced to 46.87 ± 2.03,   Effects of TRP3, TRP6, TRP9, and TRP12 on the cell viability, nitric oxide (NO) production (B), and endothelin-1 (ET-1) secretion (C) of HUVECs. All data are presented as the mean ± SD of triplicate results. ### p < 0.001, ## p < 0.01, and # p < 0.05 vs. Control; ***p < 0.001 vs. Norepinephrine (NE).

Influences of TRP3, TRP6, TRP9, and TRP12 on ROS level in H 2 O 2 -induced HUVECs
During oxidative stress, excessive ROS damage various cellular components and further induce apoptosis due to DNA damage, mitochondrial membrane potential reduction and enzyme inactivation (2,50). In the model group, increased fluorescence intensity and area after DCFH-DA staining indicated a remarkable increase in ROS content of H 2 O 2 -induced HUVECs (Figure 8). Moreover, fluorescence area and intensity decreased with the increase of ACEi peptide concentration, demonstrating that TRP3, TRP6, TRP9, and TRP12 had significant ability to decrease the ROS content in H 2 O 2 -induced HUVECs. Figure 9 quantitatively determined the ability of TRP3, TRP6, TRP9, and TRP12 to decrease the ROS content in the H 2 O 2 -induced HUVECs. At 200 μM, the ROS levels of TRP3, TRP6, TRP9, and TRP12 groups were significantly decreased from 146.6 ± 3.81% to 128.7 ± 5.2%, 127.5 ± 4.32%, 122.4 ± 4.47%, and 123.7 ± 3.21% control, respectively. Therefore, ROS levels in H 2 O 2 -induced HUVECs were significantly decreased by pretreatment of TRP3, TRP6, TRP9, and TRP12.

Production of ACEi peptides from Skipjack tuna roes
BPs released from food proteins may have high ACEi ability and alleviate cellular oxidative damage, which are two key ways to treating hypertension (17, 21). BPs hide in parent proteins and can be released by chemical degradation, proteinase hydrolysis, and microbiological fermentation methods (20,33). Proteinase hydrolysis is a popular process because of its multiple advantages, such as easily controlling the process, no pollution to the environment, and no toxic chemical residues (51,52). Because of the specificity of proteases, proteins can produce multiple hydrolysates with diverse bioactivities. Therefore, endonuclease, exonuclease, and their combinations serve as tools to yield BPs from marine creatures and their byproducts (53,54). In the experiment, we employed six proteases to hydrolyze roe proteins of Skipjack tuna, and the hydrolysate generated by flavourzyme exhibited the highest ACEi capacity (Figure 2), which further proved that the specificity of proteases can greatly affect the kind of peptides in hydrolysates, which were closely correlated with their physiological and pharmacological functions.
Protein hydrolysates are made up of diverse peptides with different MWs and physicochemical properties because of the differences in amino acid composition, which are major factors determining the methods of peptide separation (55). Large polypeptides difficultly access the molecular pockets of ACE and combine with its active sites, leading to a decrease of inhibitory ability (21, 37). For this reason, ultrafiltration and gel permeation chromatography are popularly applied to collect and isolate BPs with short chains from protein hydrolysates, such as Mustelus (14), Cyclina sinensis (56), Antarctic krill (18), tuna frame (57) and milts (44), Ulva prolifera (58), and Arthrospira platensis (59). In addition, BPs have differences in ion exchange and polarity ability due to the polar groups of amino acids, such as amino and carboxyl groups. Thereby, ion exchange chromatography and  Effects of TRP3, TRP6, TRP9, and TRP12 (100 and 200 μM) on ROS levels in H 2 O 2 -induced HUVECs. All data are presented as the mean ± SD of triplicate results. ### p < 0.001 vs. Control; ***p < 0.001 vs. Model.
Frontiers in Nutrition 12 frontiersin.org RP-HPLC are also known as common techniques for peptide purification (53,60). According to these literatures, we designed the separation process of tuna roe hydrolysate and four ACEi peptides, including TRP3 (WGESF), TRP6 (IKSW), TRP9 (YSHM), and TRP12 (WSPGF), were prepared and showed significant ACEi ability.  Peptide size and amino acid sequence are two key factors affecting the ACEi ability of BPs (17, 21). Peptide size decides the affinity between BPs and ACE because big polypeptides cannot be accommodated in and access the narrow channel of ACE (21). For example, VPP and IPP can easily access and bind the Zn 2+ of ACE channel, but ALPMHIR has low binding scores with ACE because it uncomfortably accesses the narrow channel (19). In the experiment, TRP3, TRP6, TRP9, and TRP10 are tetrapeptides or pentapeptides, and small size increases their access to the active channel of ACE, and this has been proved that the affinities of TRP3, TRP6, TRP9, and TRP12 with ACE were −8.590, −9.325, −9.703, and −8.036 kcal/mol, respectively, which were similar to those of ACEi peptides, such as SHGEY (−8. Amino acids are another factor affecting the ability of ACEi peptides. Chen et al. (64) reported that hydrophobic amino acids (Met, Ile, Phe, Trp or Lys) could significantly contribute to the inhibitory potency of peptide fraction from bighead carp (Aristichthys nobilis) hydrolysates. Moreover, the C-and N-terminal amino acids are believed to play a crucial function in the activity of ACEi peptides (21). Yu et al. (56) reported that C-terminus (such as Trp, Tyr, Pro, or Phe) and N-terminus (such as Pro, Phe, Trp, or Met) hydrophobic amino acid residues have positive effects on the ACEi activity of BPs, and the ACEi activity of WPMGF is due to the Phe and Trp residues at C-and N-terminus. Su et al. (67) found that the presence of aromatic and hydrophobic amino acids at the C-and N-terminus could significantly enhance the ACEi ability of PPLLFAAL. Wang et al. (65) concluded that YSK (IC 50 : 76 μM) and YPK (IC 50 : 38.7 μM) showed better ACEi activity than KFYG (IC 50 : 90.5 μM) and ACKEP (IC 50 : 126 μM) were due to the same N-and C-terminal amino acids (Tyr or Lys). Suo et al. (17) reported that C-terminus (Lys, Pro, or Phe) and N-terminus (Ile or Tyr) amino acids were particularly critical for the ACEi activity of IK, YEGDP and WF. Therefore, the N-and C-terminal amino acids of TRP3 (Phe and Trp), TRP6 (Ile and Trp), TRP9 (Tyr and Met), and TRP12 (Trp and Phe) are particularly important for their ACEi activity.

Cytoprotection of TRP3, TRP6, TRP9, and TRP12 on HUVECs
ECs are a constitutive part of the heart and vasculature and generally believed that ECs activation and dysfunction are preliminary processes in the pathological processes of CVDs including high blood pressure, AS and heart failure (16). Then, HUVECs commonly serve as model cells for illustrating the mechanism and developing new drugs for CVDs. In addition, EC proliferation is vital for forming new vessels, and ECs also serve as the therapeutic target of CVDs (68). Figure 8 indicated that TRP3, TRP6, TRP9, and TRP12 had no significant toxicity to HUVEC at 100 and 200 μM, which affirmed that they were appropriate for application in health products treating CVDs at concentrations below 200 μM.
NO refers to the most potent vascular endothelium factor and the deficiency of NO will raise the risks of CVDs and AS in pathologic situations. Improvement of endothelial NO production represents FIGURE 12 Effects of TRP3, TRP6, TRP9, and TRP12 on apoptosis rates of H 2 O 2 -induced HUVECs. GSH (200 μM) served as the positive control. All data are presented as the mean ± SD of triplicate results. ### p < 0.001 vs. Control; ***p < 0.001, **p < 0.01, and *p < 0.05 vs. Model.  (70), exerted their anti-hypertensive functions via memorably weakening ET-1 generation. The present results demonstrated that TRP3, TRP6, TRP9, and TRP12 could effectively protect ECs, and the mechanism was concerned with improving the level of NO, weakening the generation of ET-1, and combating the negative impact of NE on NO and ET-1 production in HUVECs.

Frontiers in
Oxidative stress represents the primary inducement of endothelial dysfunction, and it further leads to injuring the barrier function of vascular endothelium and the pathogeny of AS, hypertension and other CVDs. Additionally, high ROS content badly harms a number of functioning cell components, lowers membrane potential, deactivates antioxidant enzymes, and even results in transgene, which causes HUVECs to undergo apoptosis (47). Therefore, oxidative stress is the primary determinant for EC activation and dysfunction, and apoptosis is another main mechanism of EC injury caused by oxidative stress (17). Therefore, we established the oxidative damage model of HUVECs using 300 μM H 2 O 2 for exploring the protective capacity and mechanisms of TRP3, TRP6, TRP9, and TRP12 on EC oxidative injury.
To keep the cells in tip-top condition, the antioxidative defense system will get started timely to remove excessive ROS (70,71). MDA is a key peroxidation metabolite of the cell membrane lipid and acts as a proverbial referent to evaluate the degree of oxidative damage (55,72). ACEi peptides of IVTNWDDMEK and VGPAGPRG from Volutharpa ampullacea perryi can dose-dependently regulate NO and ET-1 generation and protect HUVECs against H 2 O 2 -induced injury, and mechanisms indicate that IVTNWDDMEK and VGPAGPRG can up-regulate the expression of Nrf2 and HO-1 to reduce the accumulation of ROS and MDA (70). FNLRMQ from Takifugu bimaculatus can be used as a potential candidate compound for alleviating the viability and apoptosis of Ang-II-induced HUVECs by regulating Nrf2/HO-1 and PI3K/Akt/eNOS signaling pathways (73). FEIHCC and EMFGTSSET from Isochrysis zhanjiangensis can alleviate endothelial damage by blocking inflammation and apoptosis of HUVECs, and mechanisms demonstrate that EMFGTSSET can regulate MAPK/NF-κB/Akt signal pathways to reduce the ROS and related cytokines (74,75). The current results turned out that the protective functions of TRP3, TRP6, TRP9, and TRP12 to H 2 O 2 -damaged HUVECs were parallel to those BPs, and their mechanisms were connected with activating Nrf2 pathway to reduce the oxidative stress level and apoptosis rate of H 2 O 2 -induced HUVECs.

Conclusion
In summary, fifteen peptides were purified from the roe hydrolysate of Skipjack tuna generated by employing flavourzyme, and four peptides with remarkable ACEi ability were identified as WGESF, IKSW, YSHM, and WSPGF, respectively. WGESF, IKSW, YSHM, and WSPGF displayed remarkable hypotensive effects via inhibiting ACE activity and regulating NO and ET-1 production in HUVECs. In addition, WGESF, IKSW, YSHM, and WSPGF could lower the oxidative stress damage and apoptosis rate of H 2 O 2 -damaged HUVECs by increasing the levels of SOD, GSH-Px, and NO to decrease the production of ROS and MDA. Therefore, this study is not only to develop a technology for the production of novel ACEi peptides of skipjack tuna roes, but also to be conducive to dealing with the problem of environmental pollution induced by tuna byproducts. Another even more important is that WGESF, IKSW, YSHM, and WSPGF might be used as natural functional ingredients for developing noticeable hypotensive products to ameliorate hypertension and CVDs. In addition, the mechanisms of WGESF, IKSW, YSHM, and WSP GF for ameliorating hypertension and cardiovascular diseases will be further investigated by in vivo experiments.

Data availability statement
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.