Production, characterisation, and biological properties of Tenebrio molitor-derived oligopeptides

Food Chemistry 450 (2024) 139400 Available online 16 April 2024 0308-8146/© 2024 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). Production, characterisation, and biological properties of Tenebrio molitor-derived oligopeptides Teresa Gonzalez-de la Rosa a , b , Sergio Montserrat-de la Paz a , b , * , Fernando Rivero-Pino a , b a Department of Medical Biochemistry, Molecular Biology, and Immunology, School of Medicine, University of Seville, 41009 Seville, Spain b Instituto de Biomedicina de Sevilla, IBiS, Hospital Universitario Virgen del Rocio/CSIC/University of Seville, 41013 Seville, Spain ARTICLE INFO Keywords: Antioxidant activity Anti-inflammatory activity Tenebrio molitor Bioactive peptide Protein hydrolysate Digestion-derived products Molecular docking ABSTRACT Three protein hydrolysates from Tenebrio molitor were obtained by enzymatic hydrolysis employing two foodgrade proteases (i.e. Alcalase and Flavourzyme), and a complete characterisation of their composition was done. The digestion-derived products were obtained using the INFOGEST protocol. In vitro antioxidant activity and anti-inflammatory activities were evaluated. Tenebrio molitor flour and the protein hydrolysates showed a high ability to scavenge the DPPH radical (EC 50 values from 0.30 to 0.87 mg/mL). The hydrolysate obtained with a combination of the two food-grade proteases could decrease the gene expression of pro-inflammatory genes after being digested. Furthermore, the peptidome was fully determined for the first time for T. molitor hydrolysates and digests, and 40 peptides were selected based on their bioactivity to be evaluated by in silico tools, including prediction tools and molecular docking. These results provide new perspectives on the use of edible insects as sustainable and not nutritionally disadvantageous food for human consumption. 1. Introduction A diet rich in vitamins, minerals, and macronutrients (proteins, fatty acids, and carbohydrates) that is well balanced helps the body avoid the development of diseases, control metabolism, and improve general health. Understanding the close relationship between nutrition and health emphasises how crucial it is to make educated dietary choices to support a robust and healthy body (Luvi´ an-Morales, Varela-Castillo, Flores-Cisneros, Cetina-P´ erez, & Castro-Eguiluz, 2022). In addition, the environmental factor that the production of food causes has recently been added to the equation, implying an increase in the complexity of the issue of healthy habits in humans regarding diet. The increase in world population estimated for the next decades, together with the environmental problems of the current food system, calls for a deep restructuration of the food system, including new sources of macronutrients to supply the food demand in a sustainable way, without compromising the health of human beings (Berardy & Sabat´ e, 2023). This situation has proven to be a key problem in the problem of climate change that humanity is currently facing, since the structure of the current system does not contribute positively to the goals set to become a carbon neutral, sustainable, and resilient society to climate change (Clark et al., 2020; Crippa et al., 2021). Furthermore, there is an increasing interest in searching for nutrients that prevent the development of diseases. In this direction, consumers are increasingly drawn to health-promoting foods as awareness of the link between nutrition and overall well-being continues to increase. In a world where lifestyles have become fast-paced, there is a growing interest in foods that not only satisfy taste buds but also contribute to optimal health. Nutrient-dense options, rich in vitamins, minerals, and antioxidants, are gaining popularity as consumers seek to support their immune systems (Anal, Koirala, Karna, Umar, & Thapa, 2023). In recent years, the use of insects as food has grown exponentially in Western societies because of their safety and nutritional composition, and considering their environmentally friendly production process, thus it has attracted the interest of the food industry as well as of the consumers. Dried yellow mealworm (Tenebrio molitor) was the first species to obtain the authorisation in the European Union (Precup et al., 2022), and consequently, research on the benefits that its consumption could exert is being carried out to fully explore and utilize this insect as food. However, public acceptance and perception of consuming insects as food can vary widely across different regions and populations. While some cultures view insects as a viable and ethical food source, others * Corresponding author at: Department of Medical Biochemistry, Molecular Biology, and Immunology, School of Medicine, Universidad de Sevilla, Av. Dr. Fedriani s/n, 41009 Seville, Spain. E-mail address: [email protected] (S. Montserrat-de la Paz). Contents lists available at ScienceDirect Food Chemistry journal homepage: www.elsevier.com/locate/foodchem https://doi.org/10.1016/j.foodchem.2024.139400 Received 23 February 2024; Received in revised form 11 April 2024; Accepted 15 April 2024 Food Chemistry 450 (2024) 139400 2 may perceive them differently, impacting their acceptance in mainstream diets. Cultural factors such as traditions, religion, and personal preferences influence attitudes toward insect consumption, highlighting the importance of understanding and addressing varying perceptions when incorporating insects in human diets (Wendin & Nyberg, 2021). From a nutritional perspective, eating insects is interesting due to their high-quality protein content (40–60% in dry basis), fatty acid profile (20–35%), vitamin and mineral content, as well as fibre in the form of chitin (10–20%). Furthermore, studies on the bioactive peptides that are encrypted in some insect species’ protein sequences are still in progress. These peptides may have a variety of functions that are relevant to human health (van Huis, Rumpold, Maya, & Roos, 2021). It has been shown that peptides derived from food proteins, generally between 2 and 20 amino acids, have bioactive properties due to their similarity with the structure of human regulatory peptides, since they can interact with some enzymes and receptors involved in human metabolism (Galland, de Espindola, Lopes, Taccola, & Pacheco, 2022). These peptides could be released from proteins by enzymatic hydrolysis using food-grade proteases, producing what are called protein hydrolysates (Rivero-Pino, 2023). Furthermore, proteins and their products generally have regulatory effects on the activation of innate and adaptive immunity (Tourkochristou, Triantos, & Mouzaki, 2021), which can depend on the amino acid residues. Peptide stability against proteases, transport and distribution capabilities, and ultimately bioactivity are determined by their structure; however, interactions between peptides and other food components, as well as intrinsic factors such as oral chewing, drug use, or genetics, can affect the toxicokinetics of a peptide (Picariello et al., 2023; Wang, Zhang, Sheng, Chi, & Wang, 2024). For Tenebrio molitor peptides obtained by enzymatic hydrolysis, bioactivities such as antioxidant, antidiabetic, or anti-inflammatory (Nongonierma & FitzGerald, 2017; Zieli´ nska, Baraniak, & Kara´ s, 2018), have been described by numerous in vitro studies, in which the identification of sequences has hardly ever been carried out. In fact, according to Teixeira, Villa, Costa, Ferreira, and Mafra (2023), only 27 sequences have been identified in T. molitor, from which only the activity of a few of them was evaluated in vitro with synthetic peptides. Consequently, there is a gap in research on the peptidome of T. molitor protein hydrolysates obtained by different enzymatic treatments, which could help unravel the mechanisms by which these pools of peptides exert certain bioactivities. In addition, studies dealing with the immunomodulatory properties of edible insect components through in vitro studies, cell-based assays, and in vivo studies (i.e. animal and human) have recently been reviewed (Rivero-Pino, Gonzalez-de la Rosa, & Montserrat-de la Paz, 2024). On top of the changes in protein structure after its hydrolysis, there is currently little information available on whether protein hydrolysis might impact the content and composition of the rest of the macronutrients of the insect meal (i.e. fibre and fat). We hypothesise that subjecting a high protein content T. molitor flour to enzymatic treatments will produce hydrolysates with improved bioactivity, while maintaining macronutrient composition. In this regard, the aim of this work was to unravel the potential of Tenebrio molitor protein hydrolysates obtained by two commonly used food-grade proteases as functional ingredients with potential health-promoting activities. To this purpose, a complete characterisation of the samples was carried out, and the bioactivity (antioxidant and anti-inflammatory) of the products after digestion was evaluated. In addition, the peptidome characterisation and evaluation through different in silico analyses including prediction tools and molecular docking of hydrolysates and their digests were also carried out. 2. Material and methods 2.1. Chemicals and samples Tenebrio molitor flour (TMF) was kindly donated by InsektLabel Biotech (Biscay, Spain). Alcalase 2.4 L FG (2.4 AU-A/g) and Flavourzyme (1000 L, 1100 LAPU/g) were kindly donated by Novozymes (Bagsvaerd, Denmark). All chemicals (reagents and solvents) were of analytical grade and provided by Sigma Chemical Co., Bachem AG (Bubendorf, CH, EU), and Gibco (Waltham, MA, USA). 2.2. Hydrolysis of Tenebrio molitor flour The hydrolysis of TMF was carried out in a jacketed reactor under continuous stirring at constant temperature (50 ◦C) and pH 8, except for the sample hydrolysed also with Flavourzyme. Alcalase-hydrolysed samples were monitored by the pH-stat method (Camacho, Gonz´ alez- Tello, P´ aez-Due˜ nas, Guadix, & Guadix, 2001) allowing to determine the degree of hydrolysis (DH). The three Tenebrio molitor hydrolysates (TMHs) were produced by resuspending TMF at 28 g of protein/L in distilled water and subjecting them to the treatments described: i) TMH10A, Alcalase was added at an enzyme/substrate (E/S) ratio of 0.375% and deactivated when DH was 10%; ii) TMH20A, Alcalase was added at an E/S ratio of 2.75% and deactivated when DH was 20%; and iii) TMH20AF: Alcalase was added at an E/S ratio of 2.75%. Once DH 20% was achieved, the pH was changed to 6 using HCl, and Flavourzyme was added at an E/S ratio of 2.75%. The reaction was deactivated after 30 min, without monitoring. Enzyme deactivation for each sample was achieved by freezing at −80 ◦C and then the samples were freeze-dried and stored until analysis. 2.3. Chemical characterisation of samples 2.3.1. Macronutrient composition Protein concentration was determined by elemental microanalysis, using a LECO TRUSPEC MICRO analyser (Leco Corporation, St. Joseph, MI, USA) and expressed as a percentage after multiplying nitrogen content by a conversion factor of 4.76 (Janssen, Vincken, van den Broek, Lakemond, & Fogliano, 2017). The fat content of the samples was determined gravimetrically as previously described with slight modification (Drusch et al., 2012) after extraction with a solvent mixture hexane/isopropanol (1:1; v:v). Total dietary fibre was determined using ANKOM 2000 technology. 2.3.2. Determination of amino acid composition For this, the samples (4–6 mg of protein) were hydrolysed by incubation with 6 N HCl at 110 ◦C for 24 h in closed tubes and under a nitrogen atmosphere. Amino acids were analysed using a Thermo Scientific liquid chromatography system consisting of a binary UHPLC Dionex Ultimate 3000 RS, connected to a quadrupole-Orbitrap QExactive hybrid mass spectrometer (ThermoFisher Scientific, USA) with HESI ionization probe. Xcalibur software was used for instrument control and data acquisition. Separation was carried out using an Acquity UPLC BEH Amide column (1.7 μ m particle size, 100 ×2.1 mm) (Waters) at 35 ◦C at a flow rate of 0.4 mL/min. A binary gradient, consisting of (A) ACN:H 2 O (85:15, v/v) and (B) H 2 O both containing 10 mM ammonium formate and 0.15% formic acid, was used with the following elution profile: 0% B (6 min), linear gradient from 5.9% B to 17.6% B (4 min), linear gradient to 29.4% B (2 min) and finally 0% B (6 min). The injection volume was 5 μ L. A data dependent acquisition method (Top5) was used in positive mode at resolution 70,000 and 17,500 at m/z 200 FWHM for Full Scan and Product Ion Scan, respectively. HESI source parameters were, spray voltage, 3.5 kV; S-lens RF level, 50; capillary temperature, 320 ◦C; sheath, auxiliary and sweep gas flow, 50, 1,3 and 3, respectively (arbitrary units); and probe heater temperature, 425 ◦C. Trace Finder 5.1 software was used for data treatment. The identification was made by comparing retention time, the exact masses of the pseudomolecular ion and their fragment ions (maximum deviation of 5 ppm) with the data contained in an amino acid compounds database with 23 compounds. Isotopic pattern scores higher than 80% were also required. T. Gonzalez-de la Rosa et al. Food Chemistry 450 (2024) 139400 3 2.3.3. Fatty acid profile An Agilent 6890 N gas chromatograph equipped with a flame ionization detector (FID) was used to separate each of the fatty acid profile and to determine its quantitative and qualitative percentages. An HP-88 capillary column of 100 m length, 0.25 mm internal diameter, and 0.2 μ m phase thickness was used as stationary phase (Sukhija & Palmquist, 1988). The identification of fatty acids was carried out by comparing their retention times with those of a mixture of standards. 2.3.4. Determination of the total phenol content Total phenol content was quantified in each sample according to the Folin−Ciocalteu spectrophotometric method. Briefly, 20 μ L of each sample at 10 mg/mL, 80 μ L of 0.7 M Na 2 CO 3 and 0.1 mL of Folin−Ciocalteu phenol reagent (0.2 M) were added and mixed. After 10 min, the absorbance was measured at 655 nm. The results were expressed as mg of gallic acid equivalent (GAE) per 100 mg of dry flour or hydrolysate (mg of GAE/100 mg sample). Each sample was analysed in quadruplicate. 2.3.5. Mineral content A microwave-assisted digestion procedure was carried out to quantify the mineral content of the samples. Reagents used for digestion were HNO 3 and H 2 O 2 , with a temperature of 200 ◦C. The analysis was performed using inductively coupled plasma optical emission spectroscopy (ICP-OES) (Santos et al., 2019). Each sample was analysed in triplicate. 2.3.6. Ultrastructural characterisation Ultrastructural characterisation of the TMF and the TMHs was done using Scanning Electron Microscopy as described in Montserrat-De la Paz et al. (2023). Images were obtained with a Zeiss Crossbeam 550 scanning electron microscope (Zeiss, Madrid, Spain) at an accelerating voltage of 2.00 kV and different magnifications. 2.4. INFOGEST: A standardised static in vitro digestion model Simulated gastrointestinal digestion (SGID) was carried out following the INFOGEST protocol (Brodkorb et al., 2019) with slight modifications to the samples (i.e. TMF, TMH10A, TMH20A, and TMH20AF). The purpose of this SGID is to simulate the effect of the enzymes digesting food components: α -amylase (EC 3.2.1.1), pepsin (EC 3.4.23.1), pancreatic lipase (EC 3.1.1.3), pancreatin, and bile salts diluted in simulated salivary, gastric, and intestinal fluid, respectively (according to the supplier’s protocol). For this purpose, 2 g of the samples were diluted 1:2 and subjected to the different phases: oral, gastric, and intestinal, for 0.5, 2–3, and 2–3 h, respectively. At the end of the SGID, enzymatic deactivation was achieved for each sample by freezing at −80 ◦C and was then freeze-dried and stored until analysis. The digested samples were encoded as dTMF, dTMH10A, dTMH20A, and dTMH20AF. 2.5. Molecular weight profile The molecular weight (MW) distribution was analysed by size exclusion chromatography (SEC) using an AKTA FPLC system. The samples were dissolved in 50 mM TrisHCl buffer pH 8 and centrifuged for 20 min at 7000g at 4 ◦C and the supernatant was filtered with a 0.2 μ m syringe filter. 500 μ L of each sample were loaded into the HiLoad 16/ 60 75 pg column pre-equilibrated with the same buffer. The absorbance was measured at 280 nm. The column was calibrated with the following standards: vitamin B12 (1.3 kDa), chemotrypsinogen (29 kDa), ribonuclease A (13.4 kDa), and bovine serum albumin (66 kDa). 2.6. Antioxidant activity The 2,2-diphenyl-1-picrylhydrazyl (DPPH, Sigma-Aldrich, Cat. No. D9132) radical scavenging activity was determined as described by Picot et al. (2010) with slight modifications. For this purpose, a 1:1 mixture containing hydrolysate, at increasing protein concentrations (0.25–5 mg/mL), and 0.1 mM DPPH in methanol was shaken and stored for 0.5 h at 25 ◦C in the dark. Then, the absorbance of the reaction mixture was measured at 515 nm. The DPPH scavenging activity was calculated from the following equation: DPPH reduction (%) = (1 - Asample −Asamplecontrol Ablank )⋅ 100 where the control sample was prepared using methanol instead of DPPH, and the blank solution contained a 1:1 mixture of DPPH and distilled water. The half-maximum effective concentration (EC 50 ) value was determined as the concentration of hydrolysate that reduces the DPPH activity by 50%. 2.7. Cell viability CACO-2 cells (ATCC-HTB-37) were incubated with different concentrations of dTMF and dTMHs in 96-well plates (10 5 cells/well) for 24 h. The concentrations used increased from 10 μ g/mL to 1 mg/mL. Subsequently, the solution of 3-(4,5-dimethylthiazol-2-yl)-2,5-diphe- nyltetrazolium bromide (MTT, Sigma Aldrich, Cat. No. M5655) at 0.5 mg/mL was added and incubated about 3 h until a purple precipitate was visible. MTT crystals were solubilised with dimethylsulfoxide and subsequently measured with a microplate reader at 570 nm corrected at 650 nm (Wu et al., 2023). 2.8. CACO-2 culture and treatments with dTMF and dTMHs CACO-2 cells were maintained in Dulbecco’s Modified Eagle’s Medium with high glucose content, supplemented with 10% heatinactivated foetal bovine serum and 1% penicillin/streptomycin. They were incubated in an atmosphere with 5% CO 2 at 37 ◦C and given fresh medium every 2–3 days. The experiments were carried out in 12-well plates, seeding CACO-2 cells at a density of 5 ×10 5 cells per well. CACO-2 cells were pre-stimulated with lipopolysaccharides (LPS, Sigma- Aldrich, Cat. No. L2880) at 1 μ g/mL. After 1 h, cells were also treated for 24 h with dTMF and dTMHs at 10 μ g/mL. 2.9. RNA isolation and RT-qPCR Total RNA was extracted using gTPzol (CViral) from LPS-stimulated CACO-2 cells. An A 260 /A 280 ratio was used in a NanoDrop ND-2000 spectrophotometer (Thermo Scientific, Madrid, Spain) to test the RNA quality. Subsequently, RNA (1 μ g) was subjected to reverse transcription (iScript, Bio-Rad, Madrid, Spain). A concentration of 2 ng/ μ L of the resulting cDNA was used as a template for real-time PCR amplifications. mRNA levels of specific genes were determined in a CFX96 system (Bio- Rad) that contained primer pairs for any gene or for glyceraldehyde 3- phosphate dehydrogenase (GAPDH) and hypoxanthine phosphoribosyltransferase 1 (HPRT) as housekeeping genes. All amplification reactions were carried out quadriplicate with average threshold cycle (Ct) numbers of the magnitude of the change in mRNA expression of candidate genes, quantified with the standard 2 –(ΔΔCt) method. All data were normalised to the endogenous reference gene (HPRT and GAPDH) level and expressed as a relative value of the control. The primers used for the amplifications are detailed in the Supplementary material - Table 1. 2.10. Peptide extraction, purification, and sequence identification Samples (TMF, TMHs, and digested ones) were acidified with 0.5% trifluoroacetic acid (v/v). The desalting and concentration step was performed with ZipTip C18 (Millipore, catalog number ZTC18M960) and the samples were dried in a speed-vacuum. LC-TIMS-MS/MS was T. Gonzalez-de la Rosa et al. Food Chemistry 450 (2024) 139400 4 carried out using a nanoElute nanoflow ultra-high pressure LC system (Bruker Daltonics, Bremen, Germany) coupled to a timsTOF Pro 2 mass spectrometer, equipped with a CaptiveSpray nanoelectrospray ion source (Bruker Daltonics) according to the procedure described in the patent P202230873. Further details on the methodology can be found in Montserrat-De la Paz et al. (2023). The reference library was acquired from UniProt_proteome_Tenebrio-molitor_Jan24. Unique protein peptides were set to larger than 1 and a high confidence score of −10lgP >17.50 was applied to indicate a protein accurately identified. 2.11. Bioactivity prediction in silico analysis The selected peptides (criteria: for each pair of samples (TMF or TMHs, and their respective digested), the ten most active sequences according to the PeptideRanker tool, which were found in both samples, with a molecular weight <1000 Da) were subjected to in silico analyses (prediction tools): a) ToxinPred software was used to predict hydrophobicity, amphipathicity, steric hindrance, charge and molecular weight (Gupta et al., 2013); b) PepCalc software was used to estimate the solubility, the number of residues and isoelectric point (Lear & Cobb, 2016); c) the anti-inflammatory activity of the peptides was estimated employing two independent tools: PreAIP software (Khatun, Hasan, & Kurata, 2019) and PreTP-Stack (Yan et al., 2023); d) PeptideRanker was used to classify sequences by the predicted probability to be bioactive (Mooney, Haslam, Pollastri, & Shields, 2012); e) AnOxPePred-1.0 was used to predict the antioxidant activities (Olsen et al., 2020); f) PASTA 2.0 (Walsh, Seno, Tosatto, & Trovato, 2014) was used to estimate secondary structure; g) The half-life in blood was predicted using PlifePred (Mathur, Singh, Mehta, Agrawal, & Raghava, 2018); h) SCMB3PP tool was used to predict the ability of peptides to penetrate the blood-brain barrier (Charoenkwan et al., 2022); and i) AllerTOP v. 2.0 estimated an allergenicity prediction of the sequences (Dimitrov, Bangov, Flower, & Doytchinova, 2014). 2.12. Molecular docking Molecular docking was carried out to determine the binding affinity energy of the selected peptides with the TLR4/MD2 receptor, following the same methodology as described in Millan-Linares, Rivero-Pino, Gonzalez-de la Rosa, Villanueva, and Montserrat-de la Paz (2024). 2.13. Statistical analysis All values are presented as means ±standard deviations (SD). Data were evaluated using Graph Pad Prism version 9.1.2 (San Diego, CA, USA). The statistical significance of any difference in parameters between the groups was assessed using two-way analysis of variance (ANOVA), followed by Tukey’s multiple-comparison test (DPPH analysis) or Dunnet multiple-comparison test (RT-qPCRs). When a comparison is made between each sample and its digested, the statistical significance is evaluated using an unpaired t-test. P values <0.05 were considered statistically significant. 3. Results and discussion 3.1. Chemical characterisation of TMF and TMHs 3.1.1. Proximate composition of the TMF and TMHs The proximate composition of TMF and its three protein hydrolysates (TMH10A, TMH20A, and TMH20AF) are shown in Table 1. The larvae of the T. molitor are distinguished for having a high protein content, although the range reported in this study (37–44%) was slightly below the range reported in the literature (Costa et al., 2020; Rivero-Pino et al., 2020), which could be due to the inconsistency among authors in the nitrogen-protein conversion factor. This factor should take into account the presence of non-protein nitrogen in insects, such as carbohydrates, such as chitin and chitosan, and consequently, whether the material is isolated or not. The fat content was also similar to the reported values (Oliveira et al., 2024) and decreased when the degree of hydrolysis increased, which could be related to the accumulation of ash together with the deterioration of fat during hydrolysis. Furthermore, it is notable that the fibre (encompassing lignine, cellulose, and hemicellulose) of all samples is significantly higher than the supplier data and published values (Zieli´ nska, Baraniak, Kara´ s, Rybczy´ nska, & Jakubczyk, 2015), so T. molitor could be an interesting food source rich in fibre that has benefits for human health, as generally fibre is described to act as a prebiotic for the gut microbiota. The variability in fibre content observed can be attributed to a combination of chemical breakdown of some of the fibre components, interference with analysis and loss of fibre-bound proteins. On top of that, although reporting the presence of phenolic compounds in T. molitor is scarce, the results showed relevant amounts of total phenolics in the analysed samples, so further studies should be carried out to precisely quantify the total phenolic content of yellow mealworm larvae samples, since phenolic compounds are vital in anti-inflammatory and antioxidant activities (Oh, Park, & Kim, 2022). In this line, the quantification of phenols in hydrolysates has been described to be higher than in non-hydrolysed samples because hydrolysis causes bound polyphenols to be released in a free state (Hou, Yoon, Oh, Sung, & Kim, 2022). On the other hand, the data of moisture (1.7%) and ash (4.9%) were measured by the supplier. Although not evaluated, a small increase in the ash after hydrolysis is expected because of the addition of NaOH and HCl in the process. In summary, the proximate analyses showed that both TMF and TMHs are equilibrated sources of macronutrients and are not nutritionally disadvantageous upon human consumption and, consequently, adequate for being included in our diets. Enzymatic hydrolysis had little effect on the macronutrient content of the samples, despite the conditions applied. 3.1.2. Amino acid composition Supplementary material - Table 2 shows the amino acid analysis of the original flour and the three hydrolysates (i.e. TMF, TMH10A, TMH20A, and TMH20AF). Furthermore, a comparison with the nutritional recommendations in adults proposed by the FAO/WHO/UNU was reported. Proteins have been widely described as the main nutritional component of yellow mealworm larvae, but the quality of insect proteins should be assessed through the essential’s amino acid content and digestibility. The amino acid content of the evaluated samples is comparable to the results reported for T. molitor by Zieli´ nska et al. (2015) and almost all amino acids satisfactory covered the requirements of essential amino acids. Furthermore, the notable presence of hydrophobic amino acids (such as valine and alanine) and aromatic amino acids (such as Table 1 Chemical composition of TMF and its protein hydrolysates (TMHs). Data are expressed as a percentage of dry weight and show mean ±standard deviation and those marked with different letters (a-d) are significantly different. n.a.: not analysed. Proximate composition (g/ 100 g of product) TMF (from supplier) TMF TMH10A TMH20A TMH20AF Protein content 56.3 44.35 ± 1.05 a 40.17 ± 1.55 a,b 42.33 ± 2.21 a 37.22 ± 1.77 b Fat content 29.7 35.57 ± 4.18 a 33.57 ± 3.94 a 31.44 ± 3.69 a 30.68 ± 3.61 a Fibre 4.4 22.66 ± 0.80 a 10.62 ± 0.22 d 17.27 ± 0.36 b 12.13 ± 0.25 c Phenols ND 0.59 ± 0.03 c 2.00 ± 0.15 b 2.78 ± 0.12 a 2.13 ± 0.14 b T. Gonzalez-de la Rosa et al. Food Chemistry 450 (2024) 139400 5 phenylalanine) can enhance the antioxidant properties of samples, which are of interest in the manufacturing of healthy ingredients (Cai, Hu, Wang, Chi, & Wang, 2022; Suo et al., 2022). In addition to that, the high content of threonine, histidine, glycine, and serine could also be considered of interest, since, for example, the intake of threonine could help in the treatment of various disorders of the nervous system, and histidine ingestion has been associated with a decrease in body mass index (BMI), adiposity, pro-inflammatory cytokines, and oxidative stress. In addition, glycine can support heart and liver health, and can reduce the risk of diabetes. Serine is a non-essential amino acid that plays an important role in brain function and the health of the central nervous system and is involved in immune function (G. Wu, 2009). However, the digestibility of the samples should be assessed to evaluate the impact of the intake of the T. molitor protein in humans, although the literature has shown that, in general, the values are similar to those obtained in plant-derived proteins (Precup et al., 2022). In relation to hydrolysates, it has been widely reported that enzymatic hydrolysis increases the digestibility of samples while also sometimes reduces the allergenicity of the samples; thus, the nutritional interest that these hydrolysates may have to specific groups of people must be considered. 3.1.3. Fatty acid composition of TMF and TMHs T. molitor has been described as a source of fat, both quantity and quality, including essential fatty acids that contribute to human adequate development. The fatty acid profile of TMF and TMHs is shown in the Supplementary material - Table 3. The results obtained in the TMF and hydrolysates are similar to the values reported by Lawal, Kavle, Akanbi, Mirosa, and Agyei (2021) and Martins da Silva et al. (2024), where linoleic (26.6–29.7%), oleic (35.5–37.57%), and palmitic (18.6–20.1%) acids are the most abundant ones. The content of linoleic acid in T. molitor samples is of interest, considering that this one is considered as an essential one, together with α -linolenic acid, although the content of this one is relatively low. According to Santa-María et al. (2023), oleic acid is an immunomodulator with an anti-inflammatory and antioxidant effects, and consequently, the nutritional interest of the fat fraction of the samples is also of interest for human consumption. It should be noted that enzymatic hydrolysis did not greatly affect the quality of fatty acids; therefore, while protein digestibility increases, the benefits from ingestion of T. molitor fat are maintained, especially for the unsaturated ones, which are more prone to oxidation due to external factors such as heat or aeration. However, it is noteworthy that the fat fraction depends on factors such as the insect’s developmental stage and sex, the farming conditions, and the feed (Rivero-Pino, Ververis, P´ erez- G´ alvez, & Naska, 2024). 3.1.4. Mineral content of TMF and TMHs The mineral content of TMF and TMHs is shown in Table 2. As can be observed, the samples were mainly rich in P, S, K, and Mg, and the only mineral in which large differences were observed among the samples was Na, whose increase is related to the addition of NaOH during enzymatic hydrolysis. When considering novel foods for human consumption, the amount of macronutrients should be evaluated, as well as the mineral content, as for some of them the authorities had established tolerable upper levels. In relation to the mineral content of these samples, the values are similar to those reported by Siemianowska et al. (2013) in T. molitor larvae, and the values found in the powder sample showed a high content of P and K (>700 mg/100 g), Mg (144.6 mg/100 g), while for other minerals the content was <10 mg/100 g, which was the case for Zn, Fe, Cu, and Mn. Similarly, the mineral composition (mg/ kg) of dried T. molitor larvae was also reported, showing values of K and P around 8 g/kg, Na and Mg around 2 g/kg, while Fe, Zn, Cu, and Mn ranged from 10 to 100 mg/kg (Costa et al., 2020). These values have been demonstrated to meet the recommended daily intake of some of these elements. Various micronutrients have been shown to be found in edible insects. In this regard, it has been described that the amount of these micronutrients of T. molitor, as well as other species, can vary depending on the stage of development and its diet (Nowak, Persijn, Rittenschober, & Charrondiere, 2016). Furthermore, the concentration of micronutrients in the final products is influenced by the processing techniques and processes used (Ojha, Bekhit, Grune, & Schlüter, 2021; Ververis et al., 2022). 3.1.5. Ultrastructure characterisation of TMF and TMHs The surface morphology of the TMF and TMHs observed by SEM is shown in Fig. 1. The protein presented in the original sample was observed to have degraded into small fragments after enzymatic hydrolysis, increasing with the reaction time, which led to a reduction in the particle size under the same SEM parameters. These findings are closely associated with recent studies with other protein sources such as hemp or turtlegrass, where protein hydrolysates showed the same behaviour, potentially correlated with the increased solubility and bioactivity of protein hydrolysates (Islam et al., 2021; Montserrat-De la Paz et al., 2023). 3.2. Chemical characterisation of digested samples The digested samples were obtained using the in vitro digestion model protocol (INFOGEST), which simulates the physicochemical changes that occur in the human gastrointestinal tract during the digestion process, so it is very important to know their digestibility and bioavailability if it is intended to be used as an ingredient for human health promotion. The digestion products can be absorbed by the epithelium cells and can produce changes in the response of the cells and consequently in the health and well-being of humans. For these reasons, in vitro digests are used in in vitro and in silico assays to analyse the biological effect of them, instead of the original samples (Oh et al., 2022). As shown in the Supplementary material – Fig. 1, the FPLC analysis of the MW profile of the peptides determined that all samples (flour/ hydrolysates and their respective digested ones) are mainly made up of peptides with MW below 1.3 kDa. In the case of TMF/dTMF, it is remarkable that a high proportion of peptides with low MW is found in the digested one compared to the original sample (TMF), which is expected because the original sample contains native protein, which would be prone to a high rate of hydrolysis by the proteases employed. However, it must be noted that TMF might contain high molecular weight protein, which, because of the limitations of the technique, is not Table 2 Mineral content of TMF and its protein hydrolysates (TMHs). Data are expressed as mg/kg and show the mean ±standard deviation (n =3) and those marked with different letters (a-d) are significantly different. n.d.: not detected. Minerals TMF TMH10A TMH20A TMH20AF Ca 560.71 ±0.1 a 548.33 ± 0.05 b 403.31 ± 0.09 c 456.09 ± 0.08 d Co n.d. n.d. n.d. n.d. Cr n.d. n.d. n.d. n.d. Cu 16.43 ±0.00 b 17.66 ±0.01 a 15.65 ±0.00 c 15.04 ±0.00 d Fe 57.45 ±0.00 a 56.31 ±0.00 b 47.71 ±0.00 d 48.52 ±0.00 c K 7689.41 ± 0.75 d 10,022.08 ± 1.35 a 8473.28 ± 0.73 c 9121.79 ± 2.18 b Mg 3073.26 ± 0.08 b 3238.47 ± 0.07 a 2709.92 ± 0.07 d 2899.08 ± 0.07 c Mn 7.15 ±0.00 a 6.87 ±0.00 b 6.36 ±0.00 c 5.7 ±0.00 d Na 1367.29 ± 0.14 d 13,984.3 ± 1.30 c 18,956.74 ± 1.68 a 16,618.15 ± 0.62 b Ni n.d. n.d. n.d. n.d. P 11,151.53 ± 0.33 b 12,156.53 ± 0.59 a 10,508.91 ± 0.38 d 10,625.91 ± 0.46 c S 5230.81 ± 0.12 b 5397.45 ± 0.08 a 4910.94 ± 0.23 d 4985.44 ± 0.16 c Se n.d. n.d. n.d. n.d. V n.d. n.d. n.d. n.d. T. Gonzalez-de la Rosa et al. Food Chemistry 450 (2024) 139400 6 detected by this column. The two hydrolysates with Alcalase (TMH10A and TMH20A) appear to be proportionally similar to their digested samples, but for further analysis, quantitative data on peak areas would be essential. However, it is notable that in the digested hydrolysates there are more peaks than in the original samples, which shows that the amount of lower MW peptides has increased at least partially. Furthermore, for the sample hydrolysed with Alcalase and Flavourzyme, the results (Fig. 2D) are similar between original and digested, which could be due to the fact that the sample is highly hydrolysed and the digestive enzymes (specific to amino acids such as Arg or Lys) would have little effect on further cleaving the peptidic bonds of the pool of peptides previously obtained, since Flavourzyme is an exopeptidase and the peptides in that sample are already of low MW (<5–7 aa). 3.3. Antioxidant activity In Fig. 2, the antioxidant activity (DPPH radical scavenging activity, which measures the direct quenching of the radical) of each sample (flour and the three protein hydrolysates and their digested samples) is reported. In the figure, the EC 50 value is represented. The antioxidant activity of the original sample and its digested was not significantly different, with values of 0.51 and 0.44 mg/mL, respectively, while the values obtained in the hydrolysates were always lower than those of the digested samples, which could imply a loss of bioactivity. Generally, hydrolysis of proteins increases their bioactivity, but it should be noted that other components could contribute to the activity of these samples (e.g., phenols). However, the obtained values (ranging from 0.30 to 0.87 mg/mL) can be considered adequate values compared to other studies and it should be evaluated with higher-quality studies whether the statistically significant difference would translate into a different biological response. Authors such as Rivero-Pino et al. (2020), Yong et al. (2023), and Mancini et al. (2019) have reported similar results on the importance of these hydrolytic enzymes, especially Flavourzyme, to enhance the antioxidant activity of protein hydrolysates of T. molitor. It is important to note that there is a discrepancy among researchers in the methodology for determining the DPPH radical scavenging activity, as the concentrations of reagents are sometimes not comparable, and consequently, it is not always reasonable to compare results with the literature if there is no positive control with a calculated value. Furthermore, different studies use extracts or isolates, whose bioactivity may not be comparable to that of crude samples. 3.4. Anti-inflammatory activity As Yoon, Oh, Park, Kim, and Kim (2023) and Rivero-Pino et al. (Rivero-Pino, Gonzalez-de la Rosa, & Montserrat-de la Paz, 2024) previously reported, protein hydrolysates from different edible insects, and Fig. 1. Surface characteristics of TMF (A) and protein hydrolysates TMH10A (B), TMH20A (C), and TMH20AF (D) by Scanning Electron Microscopy (SEM). SEM images were taken at Mag =50×and AV =2.0 kV. Scale marks: 100 μ m. Fig. 2. DPPH radical scavenging activity (EC 50 ) of each sample and its digested (TMF and dTMF; TMH10A and dTMH10A; TMH20A and dTMH20A; TMH20AF and dTMH20AF). Values (in mg of protein/mL) are presented as the mean of three replicates ±standard deviation. Different superscript letters indicate statistically significant differences between TMF and hydrolysates (a-b) and digested samples ( α -γ). Between each sample and its digested, the p-value is indicated to determine the statistical significance. T. Gonzalez-de la Rosa et al. Food Chemistry 450 (2024) 139400 7 specifically T. molitor, could have high anti-inflammatory properties considering the expression of several inflammatory-related genes, as well as in other models (in vitro analyses as well as animal studies). However, to understand the simulation of biological process, digestionderived products, which are those that can be absorbed by intestinal epithelium cells, such as CACO-2 cells, must be used to unravel the underlying mechanisms that explain this reported anti-inflammatory activity (Zieli´ nska et al., 2018), using LPS-activated human CACO-2 cells. At first, it was shown that the CACO-2 cell viability was maintained without significant change at the concentrations analysed (data not shown). Fig. 3A-D shows the expression of several pro-inflammatory genes such as TNF- α , IFN-γ, IL-1β, and IL-6. The pro-inflammatory genes TNF- α , IFN-γ, and IL-6 decrease their expression significantly after the exposure to digestion-derived hydrolysates, as also reported in previous studies (Yoon et al., 2023), compared to samples without treatment (LPS-treated, serving as control). The most notable activity was reported in the case of dTMH20AF, which, although showing the same statistical significance as the other samples, or at least two of them for TNF- α , IFN- γ, and IL-6, the expression value was lower. These results are consistent with those previously obtained in the analysis of the antioxidant activity, which can be related considering how reactive oxygen species, and the development of inflammation are somehow correlated. However, in the case of IL-1β, an absence of statistical significance was found for the other samples, except for dTMH20AF, which produced a slight decrease in gene expression. Generally, bioactive peptides are low-MW sequences, thus, in the case of dTMH20AF, which has been hydrolysed by Alcalase (non-specific endopeptidase), Flavourzyme (exopeptidase that cleaves di and tripeptides), together with digestive proteases (specific to Arg and Lys mainly, but not exclusively), the pool of peptides contained a higher amount of low-MW peptides, thus increasing the bioactivity compared to the other samples. Furthermore, the expression of the anti-inflammatory cytokine IL-4 was also measured, and the results obtained are shown in Fig. 3E. IL-4 expression increased significantly with exposure to dTMH10A, which did not occur in the remaining digestion-derived protein hydrolysates compared to control and LPS-treated cells. On top of that, Fig. 3F showed that all digested samples significantly reduced the expression of the TLR4 gene compared to LPS-treated control cells, indicating that they can interact with this transmembrane receptor, which plays a key role in the immune response to LPS stimuli. The results reported here suggest that the products obtained after digestion of TMF and hydrolysates could have a positive impact on the immunomodulation of CACO-2 cells, although validation with in vivo studies is required, is a preliminary result on the potential of hydrolysates of the T. molitor protein as bioactive immunomodulatory agents. 3.5. Peptidome profile The peptidomes of the TMF and TMHs and their respective digested samples (dTMF, dTMH10A, dTMH20A, and dTMH20A) were fully characterized by LC-TIMS-MS/MS and the most relevant results are shown in Table 3. The number of peptides identified in each sample ranged from 1197 (dTMH20AF) to 14,998 (TMH10A). It is noteworthy that after the simulated digestion occurred, the average peptide chain length is shorter than that of the peptides compared to that of the original samples, as was also reported in the molecular weight profile by FPLC, linked to the further hydrolysis occurring due to the digestive proteases. However, numerous gastro-resistant peptides were detected in all samples after the intestinal digestion phase, and these peptides are interesting for in vivo and in silico analyses because of their biological importance. Gastro-resistant peptides are those which were identified in both samples for each pair of samples, thus, the digestive enzymes did not further cleave these and are proposed are resistant to the gastrointestinal digestion. The structural characterisation of unidentified components derived from T. molitor and their functionality require further study, including Fig. 3. Effect of digested TMF (dTMF) and digested protein hydrolysates (dTMHs) on pro- and anti-inflammatory genes and TLR4 gene expression. CACO-2 cells were pre-stimulated (1 h) with lipopolysaccharide (LPS, 1 μ g/mL, black bar) and treated with dTMF and the different dTMHs at 100 μ g/mL for 24 h. The relative expression of (A) TNF- α , (B) IFN-γ, (C) IL-1β, (D) IL-6, (E) IL-4, and (F) TLR4 genes was measured by RT-qPCR. Values are presented as means ±SD (n =8) and those marked with “*” are significantly different versus the LPS control (* P <0.05; ** P <0.01; *** P <0.001; **** P <0.0001). Table 3 Summary of the peptidome characterisation of TMF and TMHs and their digested. Number of identified peptides Mean PCL of the identified peptides a Number of gastroresistant peptides b <1 kDa gastroresistant peptides (%) Mean PCL of the gastroresistant peptides of <1 KDa a TMF 6857 12 128 65.63 8 dTMF 3163 11 TMH10A 14,998 12 242 53.31 8 dTMH10A 1941 11 TMH20A 8337 12 2549 45.55 8 dTMH20A 9323 11 TMH20AF 8455 11 303 61.39 9 dTMH20A 1197 11 a Value rounded. b Gastro-resistant peptides are those which were identified in both samples for each pair of samples, thus, the digestive enzymes did not further cleave these and are proposed are resistant to the gastrointestinal digestion. PCL: peptide chain length. T. Gonzalez-de la Rosa et al. Food Chemistry 450 (2024) 139400 8 the peptidome and the contribution of each sequence to the bioactivity of protein hydrolysates, which can be estimated by in silico tools (Minkiewicz, Iwaniak, & Darewicz, 2024). In this report, to the author’s knowledge, a complete characterisation of the peptidome of TMF, TMHs, and digested samples is described (raw data not reported) for the first time in the literature (Teixeira et al., 2023), which is an excellent source of information for further studies. The use of in silico tools such as prediction bioactivity tools or molecular docking has been increasing in recent years, although their functionality and reliability is still very limited, as the correlation between in vitro and in vivo data with the in silico data is not always accurate. Consequently, it is usually preferred to employ different assays to suggest the bioactivity of the components Table 4 Toxicokinetics and bioactivity of the selected peptides identified in TMF and TMHs and their digested on the basis of in silico analyses. Sample Peptides Peptide ranker a PreAIP b PreTP- Stack c AnOXPred d Allerginicity prediction e Docking TLR4 f Half-life (s) g SCMB3PP h FRS CHEL Affinity (Kcal/mol) Cluster size TMF dTMF ADPLNGF 0.87 0.30 (N) Non-AIP 0.41 0.29 PA −9.60 109 795.21 827.33 NWGPEWG 0.87 0.44 (M) AIP 0.54 0.22 PA −10.50 26 834.91 797.67 SGGSDPDPF 0.82 0.30 (N) Non-AIP 0.41 0.25 PA −4.80 63 854.31 794.50 FAGDDAPR 0.64 0.27 (N) Non-AIP 0.39 0.22 PNA −5.20 32 845.51 850.00 ASGDWDGV 0.62 0.37 (L) Non-AIP 0.46 0.18 PA −3.40 21 855.21 782.86 GSGLIGGGLY 0.59 0.43 (M) AIP 0.47 0.23 PNA −12.30 28 826.71 798.67 NGDSGGPL 0.57 0.27 (N) Non-AIP 0.44 0.25 PNA −6.80 41 847.31 795.29 VDPIPIQPG 0.56 0.34 (N) Non-AIP 0.40 0.27 PNA −2.10 29 831.81 787.00 PGTGEGPW 0.54 0.32 (N) AIP 0.49 0.21 PA −6.10 55 839.71 812.29 AIAAPIAPIA 0.46 0.45 (M) Non-AIP 0.35 0.22 PNA −11.40 108 834.81 798.56 TMH10A dTMH10A GGGGGLGGGGGL 0.93 0.51 (H) Non-AIP 0.50 0.20 PNA −15.10 71 834.81 830.73 APLPQGPF 0.91 0.35 (L) Non-AIP 0.52 0.28 PNA −11.10 58 810.71 818.86 GLPGPPTGPL 0.89 0.33 (N) Non-AIP 0.54 0.27 PA −9.00 35 832.91 809.44 DVPSPPGGPL 0.84 0.39 (M) Non-AIP 0.55 0.30 PA −3.60 15 835.71 798.67 SGGLYPAPL 0.82 0.34 (N) Non-AIP 0.55 0.25 PA 1.10 4 930.21 843.50 SEGVFPDPF 0.79 0.35 (L) AIP 0.44 0.26 PNA −4.10 29 825.31 789.50 VADPINGF 0.74 0.32 (N) AIP 0.39 0.25 PNA −11.50 48 789.81 816.71 AAPAPAPA 0.68 0.29 (N) Non-AIP 0.45 0.27 PNA −5.90 154 834.81 850.14 SENGPLPM 0.66 0.34 (N) Non-AIP 0.46 0.27 PNA −9.50 101 832.51 800.71 GLIGAPIAAPL 0.65 0.48 (H) AIP 0.40 0.26 PNA −11.80 63 843.91 816.00 TMH20A dTMH2OA GPFMPGF 0.99 0.32 (N) Non-AIP 0.50 0.27 PNA −7.20 96 834.61 839.83 ADHPFIF 0.96 0.45 (M) AIP 0.49 0.29 PNA −10.30 37 832.71 777.17 GGGPGLGGGGF 0.95 0.42 (M) Non-AIP 0.49 0.22 PNA −9.30 20 834.71 842.10 WGPEWGL 0.95 0.42 (M) AIP 0.57 0.22 PNA −6.80 21 837.21 818.30 TPPMGWL 0.94 0.42 (M) AIP 0.51 0.27 PNA −6.80 47 811.01 805.17 SAPIWQF 0.93 0.41 (M) Non-AIP 0.42 0.25 PNA −7.20 57 882.41 796.67 PDMAFPR 0.93 0.41 (M) AIP 0.40 0.24 PA −4.80 69 835.71 810.83 GGGYGGGLGGGF 0.92 0.44 (M) Non-AIP 0.51 0.18 PNA −11.70 55 834.61 844.55 APVWAPGL 0.91 0.36 (L) Non-AIP 0.53 0.22 PNA −7.90 44 793.51 862.00 LDSPWSGF 0.91 0.49 (H) AIP 0.47 0.26 PNA −11.20 48 855.51 824.57 TMH20AF dTMH20AF GGGGGLGGGGGLG 0.92 0.49 (H) Non-AIP 0.48 0.20 PNA −13.90 56 834.81 824.50 GLPGPPTGPL 0.89 0.33 (N) Non-AIP 0.54 0.27 PA −9.00 35 832.91 809.44 ADPLNGF 0.87 0.23 (N) Non-AIP 0.41 0.29 PA −9.60 109 795.21 827.33 DVPSPPGGPL 0.84 0.39 (M) Non-AIP 0.55 0.30 PA −3.60 15 835.71 798.67 GGGSGIGGGGGIG 0.83 0.46 (M) Non-AIP 0.45 0.18 PNA −14.90 10 834.81 808.08 LPVPPPIP 0.77 0.31 (N) Non-AIP 0.53 0.29 PNA −8.60 42 835.11 784.00 GAGGAPPFQPA 0.75 0.39 (L) Non-AIP 0.57 0.26 PA −9.70 28 836.01 812.20 GPPPQGGPQ 0.74 0.33 (N) Non-AIP 0.55 0.29 PA −10.50 93 834.31 785.15 APGFGDN 0.72 0.30 (N) Non-AIP 0.41 0.26 PA −11.80 29 878.01 856.50 ADPDDLGQGG 0.70 0.33 (N) Non-AIP 0.37 0.26 PA −14.40 46 865.91 780.11 a The likelihood that the peptides would be bioactive was evaluated by PeptideRanker (http://bioware.ucd.ie/~compass/biowareweb), a server giving scores ranging from 0 (less) to 1 (more active). b Pre-AIP; http://kurata14.bio.kyutech.ac.jp/PreAIP/) is a tool predictor for classifying potential anti-inflammatory peptides, classifying the confidence as low, L (< 0.39) to high, H confidence (>0.47); N: negative; M: medium. c PreTP-Stack (http://bliulab.net/PreTP-Stack/) is used for the prediction of therapeutic peptides based on stacking ensemble learning (with a threshold of 0.5). d AnOxPePred tool (http://services.bioinformatics.dtu.dk/service.php?AnOxPePred-1.0) predicts free radical scavenging and ion chelating scores of peptides. e Allergenicity prediction was performed with AllerTOPv.20. (https://www.ddg-pharmfac.net/AllerTOP/). PA: probable allergen; PNA: probable non-allergen. g Half-life in blood was estimated using PlifePred (in seconds) accessible from https://webs.iiitd.edu.in//raghava/plifepred/. h SCMB3PP (http://pmlabstack.pythonanywhere.com/SCMB3PP) estimates the likelihood that a peptide crosses the blood-brain barrier (threshold of 749). All the tools were accessed in February 2024. T. Gonzalez-de la Rosa et al. Food Chemistry 450 (2024) 139400 9 (Minkiewicz et al., 2024), and validation with in vivo models is needed. The gastro-resistant peptides (10 from each sample) that have a MW below 1 kDa and rank higher on the PeptideRanker tool (Mooney et al., 2012) were selected because their probability of being bioactive peptides is higher and are more prone to enter into cells than those >1 kDa, according to the literature (Oliveira et al., 2024). 3.6. In silico analysis of the selected peptides 3.6.1. Physicochemical characteristics In silico analyses were carried out to the 40 selected peptides (the ten most active sequences according to the PeptideRanker tool, which were found in both samples (flour/protein hydrolysates and their digested, with a molecular weight <1000 Da). In the Supplementary material - Table 4, physical and chemical properties including net charge, pI, hydrophobicity, steric hindrance, amphipathicity, molecular weight, solubility, and length (evaluated by PepCalc and ToxinPred) are shown. In relation to these parameters, it has been described, for example, that higher the hydrophobicity, higher the likelihood of a peptide to be bioactive (Acquah, Di Stefano, & Udenigwe, 2018) or, also, it has been suggested that low steric hindrance values and high amphipathicity could be related to improved bioactivity, since it would stabilise the enzyme conformation (Mora, Gonz´ alez-Rogel, Heres, & Toldr´ a, 2020). It is interesting to note that, except for ADPDDLGQGG, all of the peptides were predicted to have 100% coils, which could be an interesting predictor of peptides to be bioactive. 3.6.2. Toxicokinetics and bioactivity The probability of being bioactive, the predicted antioxidant and anti-inflammatory activity, the probability of being an allergenic peptide, together with the estimated half-life in blood, and the ability to penetrate the blood-brain barrier of the selected peptides are shown in Table 4. In the field of bioactive peptides, one of the most widely used tools is Peptide Ranker, from which, according to the developers of the Peptide Ranker tool, a score of >0.5 (ideally >0.8) strongly suggests that the peptide is bioactive (Mooney et al., 2012). Taking into account the criteria, the threshold of PeptideRanker for TMF was 0.46, while for the hydrolysates it was 0.65 (10 A), 0.91 (20 A), and 0.70 (20 AF), in these cases, always higher than the threshold indicated by the tool developers, and especially in the case of 20 A. However, note that this analysis does not take into consideration the relative amount of peptides in each hydrolysate; thus, a correlation of these values with bioactivity in the scope of different samples is not reliable. Despite that, this is a first indication that the hydrolysis of Tenebrio molitor meal could improve the bioactivity of the sample, as several authors have reported, in relation to different bioactivities (Nongonierma & FitzGerald, 2017; Rivero-Pino, Ververis, et al., 2024; Zieli´ nska et al., 2018). This behaviour in which a hydrolysed protein might be more bioactive than the native protein is due to the exposure of residues that might interact with different targets, whereas, when encrypted in the native protein, because of structural hindrance, those sequences are unable to interact. From the selected peptides, 22 (55%) sequences were considered high, medium, or low confidence AIP according to the Pre-AIP tool, supporting the activity shown by the peptides in the cell-based analyses. However, in the case of PreTP-Stack predictions (assuming a standard threshold of 0.5), only 11 (27.5%) sequences could be considered as anti-inflammatory peptides, whereas the rest could not. The AnOxPePred tool, which aims to identify antioxidant peptides, showed the free radical scavenging (FRS) and ion chelating (CHEL) scores of all selected peptides, ranging values from 0.35 to 0.57 and 0.18–0.30, respectively, which are in line with peptides that have shown antioxidant activity when evaluated as chemically synthesised substances (Yesiltas et al., 2022). These results indicate that the generated peptide pool might be highly bioactive. On top of that, the GLIGAPIAAPL (TMH10A) and LDPSWSGF (TMH20A) peptides had a high prediction score from both anti-inflammatory tools, with scores >0.48 from PreAIP, and also exerted a high score for the FRS (>0.47). Similarly to these two sequences, GSGLIGGGLY (TMF) and GGGGGLGGGGGLG (TMH20AF) showed the potential to be highly bioactive. Regarding the half-life in blood, values ranged from 789.81 to 930.21 s, which are similar to those generally reported for food-derived bioactive peptides and other circulating hormones (values 800–900 s) (Gülseren & Vahapoglu, 2022). Half-life depends on the presence of peptidases in plasma and the clearance of the liver and kidney (Ye et al., 2022). Finally, regarding the prediction of BBB penetrating peptides, the SCMB3PP tool revealed that all peptides would be able to cross this barrier, so they can act on human brain endothelial cells (main components of the BBB), which plays a vital role in neurodegenerative disorders such as Alzheimer’s disease. As indicated previously, the identification of T. molitor peptides has been scarcely reported; therefore, this piece of evidence described herein is relevant to the research field of edible insects (Quah et al., 2023; Teixeira et al., 2023). In this sense, Zieli´ nska et al. (2018) identified NYVADGLG, AAAPVAVAK, YDDGSYKPH, and AGDDAPR as antiinflammatory peptides, and Cho and Lee (2020) identified AKKHKE and LE as antioxidant peptides from T. molitor. In this study, the peptide VADPINGF contains the VAD residues as in the peptides previously reported, increasing the evidence of the potential to be bioactive, as well as the GLG residue, appearing in four of the peptides reported here. In the same line, the terminal APR was found in the peptide FAGDDAPR. These results suggest that peptides produced from T. molitor may have anti-inflammatory properties, yet it should be highlighted that in vivo testing is necessary to determine the actual effects. To the best of the author’s knowledge, no prior identification or characterisation of the peptides proposed herein as anti-inflammatory peptides derived from T. molitor protein hydrolysates and proven to be bioavailable and bioactive has been made. 3.6.3. Molecular docking Molecular docking was carried out with the 40 selected peptides and the results obtained are reported in Table 4. The affinity range found for peptides with the TLR4/MD2 complex was wide, but in most cases it was similar to data previously reported in other studies for peptides that interact with this receptor (Jiang et al., 2022; Yu, Wang, Shuian, Liu, & Zhao, 2023), ranging from −6.80 Kcal/mol to −14.40 Kcal/mol, indicating that these peptides can interact with this receptor complex and exert immunomodulatory activity. From the analysed peptides, one was chosen from each sample, based on the in vitro and in silico results obtained, to show (in the Supplementary material – Fig. 2) the 3D representation of the binding site and interactions with the TLR4/MD2 receptor. In addition to that, in Fig. 4 the interactions between each peptide and the interaction atoms of the TLR4/MD, which are involved, are graphically represented, supporting evidence that protein hydrolysates derived from T. molitor have immunomodulatory activity. The different types of interaction are reported in Supplementary material - Table 5 as it can be observed that most of the interactions are conventional hydrogen bonds, which according to the functional group of amino acids, is the most likely to occur and stabilise the interaction (Hubbard & Kamran Haider, 2010). An easy and affordable way to find bioactive peptides is through in silico analysis. However, in silico tools still have many disadvantages and no statement on the bioactivity of peptides should be done, but just proposed (Rivero-Pino, Millan-Linares, & Montserrat-De la Paz, 2023). Thus, the subsequent research of this work should include the chemical synthesis of the identified sequences proposed as bioactive, which could allow the confirmation of the bioactivity, its stability in storage conditions, and resistance to food processing and digestive enzyme through experimental data. 4. Conclusions Tenebrio molitor larvae represent a sustainable and promising source T. Gonzalez-de la Rosa et al.