Fibrous structuring and physicochemical properties of high-moisture meat analogues formulated with defatted edible insect proteins: Effects of species and inclusion level Authors: Yea-Ji Kim, Ji Yoon Cha, Jeong-Heon Kim, Dong Hyun Keum, Bon-Jae Gu, Yun-Sang Choi Published: 2026 ABSTRACT AbstractThe growing demand for sustainable protein sources has accelerated the development of high-moisture meat analogues (HMMAs); however, achieving both nutritional quality and fibrous structure remains challenging. This study evaluated the effects of defatted edible insect proteins (Tenebrio molitor, Protaetia brevitarsis, and Gryllus bimaculatus) at 4% and 8% inclusion in soy–gluten-based HMMAs produced by high-moisture extrusion. Insect incorporation influenced expansion, density, and color, with reduced lightness and increased redness at higher inclusion levels. T. molitor formulations showed improved protein quality, including increased methionine content and the highest essential amino acid index. These samples also exhibited enhanced intermolecular interactions (hydrophobic interactions, hydrogen bonding, and disulfide cross-linking), leading to improved structural stability. Microstructural analysis confirmed more compact and aligned fibrous structures, while texture results indicated more balanced mechanical properties at moderate inclusion levels. Overall, edible insect proteins improved HMMA properties in a species-dependent manner, with T. molitor showing the most promising performance.Keywords: Meat analogues, Extrusion, Tenebrio molitor, Protaetia brevitarsis, Gryllus bimaculatus Introduction The growing demand for sustainable protein foods has accelerated the development of meat analogues that can reduce reliance on conventional livestock production while maintaining desirable nutritional and sensory qualities (Cha et al., 2025; Lee et al., 2025). Among these, high-moisture meat analogues (HMMAs) produced by high-moisture extrusion have received considerable attention because thermomechanical processing can generate anisotropic fibrous structures and meat-like textures in protein-rich matrices (Wittek, Zeiler, Karbstein and Emin, 2021, Zhang et al., 2023). During extrusion, protein denaturation, alignment, aggregation, and phase separation under combined heat, shear, and moisture conditions collectively determine the formation of layered structures resembling muscle fibers (Schmid et al., 2022). Therefore, the molecular properties of protein ingredients are important factors affecting HMMA quality, although it is difficult to accurately predict the structuring process under HMMA processing conditions, where proteins undergo drastic denaturation and degradation (Xia et al., 2023). Most HMMAs are primarily formulated with soy, wheat gluten, peas, and other plant proteins (Schmid, Farahnaky, Adhikari and Torley, 2022, Zhang et al., 2023). However, plant-based systems are limited in terms of nutritional balance and textural diversity. Edible insects have emerged as promising complementary protein sources because of their high protein content, favorable amino acid profiles, and lower environmental burden compared to conventional animal proteins (Queiroz et al., 2023). In addition, insect proteins exhibit useful techno-functional properties, including water-binding, emulsifying, and gelation behaviors, which are directly related to structure formation in extruded foods (Mokaya et al., 2024; Queiroz et al., 2023). Despite this potential, the application of edible insect proteins in HMMAs remains limited, and their behavior in high-moisture extrusion systems is not yet well understood. Compared with plant proteins, insect-derived ingredients differ in protein composition, molecular conformation, non-protein fractions, pigments, and residual lipids, all of which may affect rheology, intermolecular interactions, thermal transitions, and fibrous structure formation during extrusion (Queiroz et al., 2023; Xia et al., 2023). Among edible insects of growing interest in Korea and other Asian markets, Tenebrio molitor, Protaetia brevitarsis, and Gryllus bimaculatus are relevant but compositionally distinct protein sources (Kim et al., 2022). T. molitor is widely recognized as an edible insect ingredient (EFSA Panel on NDA, 2021), and has been utilized in meat analogue (Kim et al., 2024). The possibility of mealworm protein as an ingredient of high-moisture extrusion has been evaluated at various extrusion temperatures (Chao et al., 2025). P. brevitarsis has attracted attention because of its favorable protein characteristics and the strong influence of defatting conditions on its functionality (Kim et al., 2021; Lee et al., 2023). The value of the protein extracted from G. bimaculatus has also been confirmed in 3D printing-based meat analogue structuring (Nam et al., 2023). However, comparative evidence of how distinct edible insect proteins behave under identical HMMA processing conditions remains limited. In addition, the level of insect protein inclusion is an important but underexplored factor. Moderate incorporation into soy–gluten systems may improve protein quality by enhancing the indispensable amino acid balance, whereas excessive inclusion may disrupt protein–protein interactions, water distribution, and network continuity, thereby impairing fibrous structure formation. Because HMMA quality is governed by fibrous structuring and physicochemical properties formed through protein interactions during extrusion (Xia et al., 2023), optimizing the inclusion level is essential to balance nutritional enhancement with structural integrity. Therefore, a systematic comparison under identical extraction conditions is needed to determine whether species- and inclusion level-dependent differences in insect proteins can enhance nutritional quality without compromising fibrous structure formation. In this study, soy-gluten-based HMMAs containing 4% or 8% defatted T. molitor, P. brevitarsis, and G. bimaculatus were produced, and their amino acid composition, physicochemical properties, intermolecular interactions, thermal characteristics, microstructure, and texture were evaluated. Materials and methods Materials Larvae of Tenebrio molitor (TM) and Protaetia brevitarsis seulensis (PB), together with adult Gryllus bimaculatus (GB), were procured from a commercial edible insect producer (Farm Bang, Jeongeup, Republic of Korea). Isolated soy protein (ISP; protein content of 90%, dry basis) and vital wheat gluten were provided by CJ CheilJedang (Seoul, Republic of Korea). Food-grade cornstarch was purchased from Daesang Co. (Seoul, Republic of Korea). Unless otherwise specified, all chemicals and reagents used for analytical procedures were of analytical grade and obtained from Sigma–Aldrich (St. Louis, MO, USA). Edible insect preparation and defatting process Fresh edible insects were rinsed thoroughly with distilled water to remove surface impurities, frozen at −80 °C, and subsequently freeze-dried for 72 h using a freeze dryer (FD8512, Ilshin BioBase, Dongducheon, Republic of Korea). The dried materials were milled using a laboratory grinder (MF10, IKA, Staufen, Germany) and passed through a 60-mesh sieve to obtain a uniform powder. Lipid removal was performed using food-grade n-hexane following a modified solvent extraction procedure described by Kim et al. (2021). Briefly, the insect powder was mixed with n-hexane at a solid-to-solvent ratio of 1:5 (w/v) and stirred at 25 °C for 6 h. After extraction, the mixture was centrifuged at 4000 ×g for 15 min, and the solvent phase was removed. The extraction cycle was repeated three times to maximize defatting efficiency. The resulting defatted powders were left under a fume hood overnight to evaporate residual solvent. The proximate compositions of the defatted insect powders were presented in Table S1. Before the manufacture of HMMAs, the insect powders were sealed and stored at −20 °C until use. Preparation of high-moisture meat analogue (HMMA) The basal formulation for HMMA production consisted of isolated soy protein, wheat gluten, and cornstarch in a weight ratio of 6:3:1. Defatted edible insect powder was incorporated at two concentrations (4% and 8%, w/w), replacing an equivalent amount of ISP. These addition levels were selected based on preliminary extrusion trials, with 8% identified as the highest feasible level for successful processing of all defatted edible insect powder samples. The dry ingredients were blended for 10 min in a ribbon mixer to ensure homogeneous dispersion. Prior to extrusion, distilled water was added to adjust the moisture content of the feed mixture to 65%. High-moisture extrusion was performed using a corotating twin-screw extruder equipped with a cooling die (IFTE-2540V2S1L1, MKE Polymer, Republic of Korea). The barrel temperature profile from the feeding zone to the die was set at 20, 70, 150, 150, and 120 °C. Screw speed and feed rate were maintained at 300 rpm and 5.4 kg/h, respectively. The cooling die temperature and length were 50 °C and 600 mm. The die pressure during high-moisture extrusion ranged from 5.97 to 11.63 bar, depending on the type and level of defatted edible insect powder (Table S2). Extrudates discharged from the cooling die were immediately collected, cut into uniform segments, vacuum-packaged, and stored at 4 °C prior to physicochemical and structural analyses. The experimental treatments were designated according to the insect species and inclusion level. CON represented the control HMMA without insect protein. TM4 and TM8 contained 4% and 8% defatted Tenebrio molitor larval powder, respectively; PB4 and PB8 contained 4% and 8% defatted Protaetia brevitarsis larval powder, respectively; and GB4 and GB8 contained 4% and 8% defatted Gryllus bimaculatus powder, respectively, replacing part of the ISP. Expansion ratio and density The transverse expansion of the extrudates was assessed after extrusion and cooling by measuring sample thickness at five randomly selected points along each HMMA strip using a digital caliper (Mitutoyo Corp., Kawasaki, Japan). The expansion ratio was calculated as the ratio of the average extrudate thickness to the die thickness. Bulk density was determined using a displacement-based approach. HMMA samples were cut into cylindrical pieces of identical length (20 mm), and their masses were recorded using an analytical balance. Sample volume was estimated by water displacement at ambient temperature, and bulk density was expressed as mass per unit volume (g/cm3). Color The surface color of HMMA samples was measured using a calibrated colorimeter (CR-400, Konica Minolta, Tokyo, Japan). Instrument calibration was performed using a standard white tile prior to analysis. Color values were recorded as CIE L* (lightness), a* (redness), and b* (yellowness) at five different positions on each sample, and mean values were used for statistical analysis. Proximate composition Moisture, crude protein, crude fat, and ash contents were determined according to standard Association of Official Analytical Chemists (AOAC) methods (AOAC, 2023). Moisture content was measured by atmospheric oven drying, where approximately 3 g of sample was heated at 100 °C until constant weight was reached. Crude protein was quantified using the Kjeldahl method with 1 g of sample, and nitrogen content was converted to protein using a factor of 6.25. Crude fat was analyzed from 2 g of sample using ether extraction following acid hydrolysis. Ash content was determined by incinerating 1 g of sample in a muffle furnace at 550 °C until complete combustion of organic matter. All results are expressed on a wet basis unless otherwise stated. Amino acid and protein quality For amino acid analysis, HMMA samples were hydrolyzed in 6 M HCl at 110 °C for 24 h under a nitrogen atmosphere. After hydrolysis, samples were cooled to room temperature, diluted with distilled water, and filtered through a 0.2 μm PTFE membrane prior to chromatographic analysis. Amino acid profiles were determined using an automatic amino acid analyzer (L-8900, Hitachi, Tokyo, Japan) equipped with an ion-exchange column (4.6 × 60 mm) packed with Hitachi-custom ion-exchange resin. Amino acids were quantified using a certified standard mixture (AAS18, Sigma–Aldrich, St. Louis, MO, USA). Detection was performed at 570 nm for most amino acids and at 440 nm for proline. The essential amino acid index (EAAI) was calculated using the indispensable amino acid reference pattern for adults established by FAO/WHO/UNU. Intermolecular forces The relative contributions of intermolecular interactions within the HMMA protein network were estimated using a sequential solvent extraction method based on differential protein solubilization. Finely minced HMMA samples (2.0 g) were homogenized separately with 10 mL of each extraction medium: 0.6 M NaCl (S1), 0.6 M NaCl containing 1.5 M urea (S2), 0.6 M NaCl containing 8 M urea (S3), and 0.6 M NaCl containing 8 M urea plus 0.5 M β-mercaptoethanol (S4). Each extraction was conducted at 4 °C for 1 h, followed by centrifugation at 10,000 ×g for 15 min. Protein concentration in the supernatant was quantified using the Bradford assay with bovine serum albumin as the standard. The relative strength of specific interactions was inferred from differences in protein solubility among solvent systems: the increase from S1 to S2 was attributed to hydrogen bonding, from S2 to S3 to hydrophobic interactions, and from S3 to S4 to disulfide bond disruption. Differential scanning calorimetry (DSC) The thermal transition behavior of the HMMA samples was analyzed using a differential scanning calorimeter (DSC4000, PerkinElmer, Waltham, MA, USA). Approximately 30 mg of each sample was weighed into an aluminum pan and hermetically sealed, with an empty sealed pan serving as a reference. Measurements were carried out under a nitrogen purge to minimize oxidative changes during heating. Samples were scanned from 20 °C to 120 °C at a heating rate of 10 °C/min. Thermograms were recorded continuously, and the peak transition temperature and enthalpy change (ΔH) were calculated using the manufacturer's software. Fourier transform infrared spectroscopy (FTIR) and secondary structure FTIR spectra were collected to examine structural changes in the protein matrix of the HMMAs. Prior to analysis, samples were freeze-dried and ground into a fine powder. Each sample was thoroughly mixed with spectroscopic-grade KBr and compressed into pellets. Spectra were acquired using an FT/IR-8× spectrometer (JASCO, Tokyo, Japan) over the range of 4000–400 cm−1 at a resolution of 4 cm−1. Fifty scans were accumulated for each spectrum to improve signal quality. Background spectra obtained under identical conditions were automatically subtracted. Protein secondary structure was evaluated by deconvoluting the amide I band (1700–1600 cm−1) using PeakFit version 4.12 (Grafiti, Palo Alto, CA, USA). Microstructure observation The microstructural characteristics of HMMAs were observed using a scanning electron microscope (S-4700, Hitachi, Tokyo, Japan). Samples were cut into approximately 0.3 × 0.3 × 0.3 cm cubes and freeze-dried to remove moisture while preserving internal structure. The dried specimens were mounted on aluminum stubs using conductive carbon tape, with the longitudinal orientation aligned to visualize the extrusion-induced fibrous architecture. After sputter coating with gold, samples were examined at an accelerating voltage of 10 kV. Images were captured at 100 × and 500 × magnifications to evaluate overall alignment and the detailed lamellar morphology, respectively. Texture profiles The textural properties of HMMA samples were measured using a texture analyzer (TA-XT Plus, Stable Micro Systems, Surrey, UK), following a previously reported procedure with minor modifications (Choi et al., 2024). Before testing, samples were equilibrated to room temperature and cut into two different dimensions depending on the analysis: 3 × 3 × 1 cm (C1) for texture profile analysis (TPA) and 1 × 1 × 1 cm (C2) for directional shear measurements. For TPA, C1 samples were compressed twice to 60% strain using a 50 mm cylindrical probe at a crosshead speed of 2 mm/s. From the resulting force–time curves, hardness, springiness, cohesiveness, gumminess, and chewiness were calculated according to standard TPA definitions. The development of the fibrous structure was further assessed by determining the texturization index using a V-blade probe. C2 samples were sheared in both the longitudinal and transverse directions relative to the extrusion-induced fiber orientation at a test speed of 2 mm/s. The texturization index was expressed as the ratio of shear forces measured in the two directions. Statistical analysis All measurements were performed in triplicate, and results are presented as mean ± standard deviation. Statistical analyses were performed using SPSS Statistics version 25 (IBM Corp., Armonk, NY, USA). Significant differences among treatments were evaluated using one-way analysis of variance (ANOVA), followed by Duncan's multiple range test for post hoc comparisons when the main effect was significant. Differences were considered statistically significant at p < 0.05. Results and discussion Expansion ratio, density, and color The effects of edible insect species and inclusion levels on the expansion behavior, bulk density, and color of high-moisture meat analogues (HMMAs) are presented in Table 1. The incorporation of defatted edible insect proteins significantly influenced these properties, although the magnitude of change in expansion ratio was relatively limited (0.96–1.03). Among the treatments, TM4 and GB4 exhibited the highest expansion ratios (∼1.02–1.03), whereas TM8 showed the lowest value (0.96), indicating that increasing the inclusion level of T. molitor reduced radial expansion. In high-moisture extrusion systems, expansion is constrained by the dominance of protein alignment and phase separation over puffing and is primarily governed by melt viscoelasticity and water vapor flash-off at the die exit (Dekkers et al., 2018; Schmid et al., 2022). Therefore, the reduced expansion observed in TM8 may indicate altered melt flow behavior possibly associated with enhanced protein–protein interactions that restrict bubble growth. Table 1Expansion ratio, density, and color of high-moisture meat analogues (HMMAs) containing edible insect proteins.TraitsCON1)TM4TM8PB4PB8GB4GB8Expansion ratio1.00 ± 0.03bc2)1.02 ± 0.01a0.96 ± 0.01d0.99 ± 0.02c1.02 ± 0.02ab1.03 ± 0.01a1.01 ± 0.01abDensity (g/cm3)1.20 ± 0.02a1.15 ± 0.02bcd1.18 ± 0.02ab1.17 ± 0.03abc1.15 ± 0.01cd1.14 ± 0.01cd1.14 ± 0.03dL*67.24 ± 0.56a58.67 ± 0.47d56.50 ± 0.34e60.58 ± 0.60c57.01 ± 0.77e61.40 ± 0.25b56.42 ± 0.89ea*1.60 ± 0.07f2.70 ± 0.16d3.10 ± 0.11b2.91 ± 0.14c3.59 ± 0.05a2.45 ± 0.08e3.15 ± 0.14bb*19.05 ± 0.74a16.15 ± 0.40c15.66 ± 0.19c17.22 ± 0.58b16.46 ± 0.87bc14.80 ± 0.58d13.50 ± 0.69e1)CON, HMMA without insect protein; TM4, HMMA with defatted Tenebrio molitor larvae 4%; TM8, HMMA with defatted Tenebrio molitor larvae 8%; PB4, HMMA with defatted Protaetia brevitarsis larvae 4%; PB8, HMMA with defatted Protaetia brevitarsis larvae 8%; GB4, HMMA with defatted Gryllus bimaculatus 4%; GB8, HMMA with defatted Gryllus bimaculatus 8%.2)Different superscript letters (a-f) in a row indicate significant differences between HMMAs (p < 0.05). Expansion ratio, density, and color of high-moisture meat analogues (HMMAs) containing edible insect proteins. CON, HMMA without insect protein; TM4, HMMA with defatted Tenebrio molitor larvae 4%; TM8, HMMA with defatted Tenebrio molitor larvae 8%; PB4, HMMA with defatted Protaetia brevitarsis larvae 4%; PB8, HMMA with defatted Protaetia brevitarsis larvae 8%; GB4, HMMA with defatted Gryllus bimaculatus 4%; GB8, HMMA with defatted Gryllus bimaculatus 8%. Different superscript letters (a-f) in a row indicate significant differences between HMMAs (p < 0.05). Bulk density ranged from 1.14 to 1.20 g/cm3 and showed a generally inverse relationship with expansion. All insect-containing samples exhibited slightly lower density than the control, although TM8 showed relatively higher density among the insect treatments, consistent with its reduced expansion. This indicates that insect proteins modulated matrix compactness, with GB and PB treatments forming less densely packed structures, whereas TM8 promoted a more consolidated network. In high-moisture meat analogues, moderate density is desirable because it reflects a compact, anisotropic structure without excessive porosity, which is essential for a meat-like texture (Dekkers et al., 2018). These results suggest that T. molitor enhances structural consolidation during extrusion, particularly at higher inclusion levels. Color parameters were strongly affected by insect incorporation. Lightness (L*) decreased significantly from 67.24 (control) to 56.42–61.40 in insect-containing samples, whereas redness (a*) increased from 1.60 to 2.45–3.59, and yellowness (b*) decreased from 19.05 to 13.50–17.22. At the lower inclusion level (4%), TM and PB showed darker and redder colors than GB. The color differences compared to the control were more pronounced at 8% inclusion across all species. The observed darkening and increased redness can be attributed to both intrinsic insect pigments and enhanced Maillard reactions under high-temperature extrusion conditions (Nogales-Mérida et al., 2019). The shift toward darker and redder tones may be advantageous for mimicking the appearance of cooked meat (Samard & Ryu, 2019); however, excessive darkening at higher inclusion levels may compromise visual acceptability. Collectively, these results indicate that insect species and inclusion levels influence not only structural properties but also product appearance, which is a critical quality attribute of meat analogues. Proximate composition The proximate compositions of HMMAs formulated with different defatted edible insect proteins are presented in Table 2. The composition was significantly affected by insect incorporation, reflecting both compositional differences among species and their interactions with the soy–gluten matrix. Although the same amount of water was added to all formulations, moisture content varied depending on the insect species and inclusion level (60.15–64.19%). This variation can be associated with differences in the water absorption and matrix-forming behavior of insect proteins, while chitin-containing fractions may also contribute to water retention depending on the insect species (Mokaya et al., 2024). Crude protein content ranged from 27.03% to 31.22%, with TM4 showing the highest value and PB4 the lowest. The lack of a consistent increase with higher inclusion levels indicates that protein content was influenced not only by ingredient composition but also by moisture redistribution and non-protein fractions. The relatively higher protein content in T. molitor treatments suggests better integration into the protein matrix, consistent with previous findings (Belluco et al., 2013). Meanwhile, the high moisture and low protein contents of HMMA with P. brevitarsis are consistent with the proximate composition of defatted P. brevitarsis protein (Table S1). In addition, the higher carbohydrate content of P. brevitarsis, assuming a high proportion of chitin, may disrupt the formation of a continuous protein network (Choi et al., 2026). The crude fat content of HMMA remained low (1.26–1.74%) across all treatments, confirming the effectiveness of the defatting process. This is important because residual lipids can act as lubricants during extrusion, reducing shear forces and hindering protein alignment and network formation (Schmid, Farahnaky, Adhikari and Torley, 2022, Zhang et al., 2023). Therefore, the low fat levels likely contributed to maintaining favorable structuring conditions. Ash content showed slight variation (1.19–1.28%) without a consistent increasing trend across all treatments. While this may enhance nutritional value, increased ash content can dilute the protein matrix and affect network continuity. This compositional shift may partly explain the differences observed in structural and textural properties among treatments. Overall, the compositional data indicate that T. molitor provided the most balanced profile in terms of protein enrichment, moderate moisture, and a limited increase in ash, which are favorable for both nutritional and structural performance. Table 2Proximate composition of high-moisture meat analogues (HMMAs) containing edible insect proteins.TraitsCON1)TM4TM8PB4PB8GB4GB8Moisture content (%)62.57 ± 0.10d2)60.15 ± 0.06f62.89 ± 0.09b64.19 ± 0.03a62.72 ± 0.02c61.84 ± 0.05e62.98 ± 0.04bProtein content (%)28.65 ± 0.46c31.22 ± 0.46a28.13 ± 0.58cd27.03 ± 0.12e27.62 ± 0.39de30.18 ± 0.05b28.90 ± 0.10cFat content (%)1.72 ± 0.08a1.65 ± 0.06a1.26 ± 0.10b1.74 ± 0.03a1.67 ± 0.00a1.62 ± 0.09a1.67 ± 0.09aAsh content (%)1.26 ± 0.03a1.28 ± 0.01a1.19 ± 0.03b1.19 ± 0.03b1.27 ± 0.01a1.22 ± 0.03ab1.23 ± 0.03ab1)CON, HMMA without insect proteins; TM4, HMMA with defatted Tenebrio molitor larvae 4%; TM8, HMMA with defatted Tenebrio molitor larvae 8%; PB4, HMMA with defatted Protaetia brevitarsis larvae 4%; PB8, HMMA with defatted Protaetia brevitarsis larvae 8%; GB4, HMMA with defatted Gryllus bimaculatus 4%; GB8, HMMA with defatted Gryllus bimaculatus 8%.2)Different superscript letters (a-f) in a row indicate significant differences between HMMAs (p < 0.05). Proximate composition of high-moisture meat analogues (HMMAs) containing edible insect proteins. CON, HMMA without insect proteins; TM4, HMMA with defatted Tenebrio molitor larvae 4%; TM8, HMMA with defatted Tenebrio molitor larvae 8%; PB4, HMMA with defatted Protaetia brevitarsis larvae 4%; PB8, HMMA with defatted Protaetia brevitarsis larvae 8%; GB4, HMMA with defatted Gryllus bimaculatus 4%; GB8, HMMA with defatted Gryllus bimaculatus 8%. Different superscript letters (a-f) in a row indicate significant differences between HMMAs (p < 0.05). Amino acid and protein quality The amino acid profiles and EAAI of HMMAs formulated with defatted edible insect proteins are presented in Table 3. Incorporation of insect proteins improved amino acid profiles and protein quality in a species-dependent manner. All formulations contained complete essential amino acid profiles, with EAAI values ranging from 1.29 (control) to 1.42 (TM8), indicating that all samples exceeded the FAO reference pattern (FAO, 2013). Notably, methionine content increased substantially from 12.65 mg/g protein in the control to 16.58–16.59 mg/g in TM formulations, representing the most significant nutritional improvement. This enhancement is particularly important because sulfur-containing amino acids are often limited in plant-based proteins. The observed increase demonstrates a strong complementary effect between insect and plant proteins, consistent with previous studies (Queiroz et al., 2023). In addition, branched-chain amino acids (BCAAs), including valine and isoleucine, were significantly increased by insect protein addition, which are relevant for muscle protein synthesis and metabolic regulation (Wolfe, 2017). Although GB in its adult stage possesses more abundant muscle proteins in well-developed organs such as the legs (Kim et al., 2026), the larval stages of TM and PB may result in a more effective accumulation of nutritionally valuable amino acids. Beyond nutritional aspects, amino acid composition may also influence structuring behavior during extrusion. Increased levels of glycine and alanine in TM formulations may enhance protein flexibility and facilitate molecular rearrangement, whereas sulfur-containing amino acids contribute to a favorable environment for covalent cross-linking. These compositional advantages are consistent with improved intermolecular interactions and structural properties in TM samples. Therefore, T. molitor, particularly at 8% inclusion, improved both protein quality and functional performance in HMMA systems. Table 3Amino acid profiles of high-moisture meat analogues (HMMAs) containing edible insect proteins.TraitsCON1)TM4TM8PB4PB8GB4GB8Essential amino acid (EAA, mg/g protein)His23.32 ± 0.3622.30 ± 0.1522.03 ± 0.7822.39 ± 1.2322.94 ± 0.2523.35 ± 0.2122.99 ± 0.35Ile35.69 ± 1.39c3)39.85 ± 0.16a40.41 ± 0.68a40.01 ± 2.14a39.66 ± 0.08ab38.49 ± 1.18ab37.00 ± 0.04bcLeu77.41 ± 1.4976.34 ± 0.4476.32 ± 2.4176.12 ± 3.6676.01 ± 1.8978.13 ± 1.0075.78 ± 0.97Lys50.22 ± 0.8353.12 ± 0.2752.03 ± 1.6253.29 ± 2.7153.82 ± 0.2650.12 ± 0.5452.15 ± 0.50Met12.65 ± 0.88c16.59 ± 0.01a16.58 ± 0.19a15.54 ± 0.58ab14.59 ± 1.33bc13.50 ± 0.09c13.48 ± 1.21cPhe + Tyr88.53 ± 1.4386.33 ± 1.3593.76 ± 2.6191.29 ± 4.5786.09 ± 0.2789.42 ± 0.2185.79 ± 2.78Thr37.94 ± 0.2539.03 ± 0.2341.47 ± 1.6739.90 ± 2.3039.85 ± 0.2038.12 ± 0.3640.33 ± 0.11Val36.69 ± 0.80d44.40 ± 0.09b47.02 ± 1.03a41.64 ± 2.02c41.99 ± 0.11c39.78 ± 0.66c40.03 ± 0.15c Non-essential amino acid (Non-EAA, mg/g protein)Ala38.41 ± 0.57c40.97 ± 0.03b44.47 ± 1.38a41.06 ± 1.90b39.69 ± 0.07bc38.44 ± 0.50c41.06 ± 0.81bArg62.23 ± 1.2062.06 ± 0.1361.85 ± 2.4161.05 ± 2.7761.20 ± 0.9463.17 ± 0.1963.55 ± 1.35Asp93.31 ± 1.21c98.69 ± 0.75ab99.92 ± 2.07a95.67 ± 4.65abc93.78 ± 0.22bc90.36 ± 0.26c92.82 ± 1.90cGlu265.33 ± 1.62a253.47 ± 2.93a228.07 ± 6.84b249.67 ± 15.42a255.97 ± 2.83a261.11 ± 0.84a257.06 ± 2.61aGly40.19 ± 0.41bc42.39 ± 0.19b46.11 ± 1.32a42.00 ± 2.18bc42.08 ± 0.15bc39.73 ± 0.13c41.15 ± 0.91bcPro80.78 ± 2.7274.49 ± 2.8176.12 ± 3.8876.74 ± 6.4878.02 ± 1.1780.26 ± 1.8381.38 ± 2.41Ser57.29 ± 0.18a49.96 ± 0.83b53.85 ± 2.08a53.63 ± 2.84a54.33 ± 1.07a56.02 ± 0.30a55.43 ± 0.50aEAA index2)1.29 ± 0.03c1.40 ± 0.00ab1.42 ± 0.03a1.38 ± 0.06ab1.36 ± 0.02abc1.33 ± 0.01bc1.33 ± 0.03bc1)CON, HMMA without insect protein; TM4, HMMA with defatted Tenebrio molitor larvae 4%; TM8, HMMA with defatted Tenebrio molitor larvae 8%; PB4, HMMA with defatted Protaetia brevitarsis larvae 4%; PB8, HMMA with defatted Protaetia brevitarsis larvae 8%; GB4, HMMA with defatted Gryllus bimaculatus 4%; GB8, HMMA with defatted Gryllus bimaculatus 8%.2)EAA index was calculated using the adult indispensable amino acid requirements of WHO/FAO/UNU expert consultation.3)Different superscript letters (a-d) in a row indicate significant differences between HMMAs (p < 0.05), whereas rows without superscript letters indicate no significant differences among treatments (p > 0.05). Amino acid profiles of high-moisture meat analogues (HMMAs) containing edible insect proteins. CON, HMMA without insect protein; TM4, HMMA with defatted Tenebrio molitor larvae 4%; TM8, HMMA with defatted Tenebrio molitor larvae 8%; PB4, HMMA with defatted Protaetia brevitarsis larvae 4%; PB8, HMMA with defatted Protaetia brevitarsis larvae 8%; GB4, HMMA with defatted Gryllus bimaculatus 4%; GB8, HMMA with defatted Gryllus bimaculatus 8%. EAA index was calculated using the adult indispensable amino acid requirements of WHO/FAO/UNU expert consultation. Different superscript letters (a-d) in a row indicate significant differences between HMMAs (p < 0.05), whereas rows without superscript letters indicate no significant differences among treatments (p > 0.05). Intermolecular forces and thermal stability The intermolecular interactions and thermal transition behaviors of HMMAs formulated with defatted edible insect proteins are presented in Table 4. Intermolecular interactions play a central role in the formation of fibrous structures in extruded high-moisture products. Hydrogen bonding increased significantly with insect incorporation, reaching the highest value in TM8 (0.15 mg/mL), suggesting enhanced protein–protein and protein–water interactions. This increase indicates conformational changes and reorganization of insect proteins, which are essential for forming a continuous anisotropic network. In addition, TM formulations maintained hydrophobic interaction levels comparable to the control. Whereas, hydrophobic interactions did not increase with increasing GB inclusion and significantly decreased as the PB inclusion level increased (p < 0.05), indicating reduced aggregation efficiency (Schmid, Farahnaky, Adhikari and Torley, 2022, Zhang et al., 2022). Disulfide bond formation, critical for stabilizing gluten-based networks, was enhanced in TM treatments and GB8, with TM4 showing the highest value. This may be attributed to the higher content of hydrophobic amino acids (Ile, Val, Ala, and Met) in TM-added HMMAs, which can contribute to more complete protein unfolding under extrusion conditions and facilitate disulfide bond cross-linking (Samard et al., 2019). In GB-added HMMAs, disulfide bond slightly increased with GB inclusion level, suggesting that the reduction in hydrogen bonding and hydrophobic interactions observed at the lower GB inclusion level was accompanied by a shift toward covalent disulfide cross-linking within the protein network. Disulfide cross-linking contributes to the mechanical strength and integrity of fibrous structures in HMMAs (Dekkers et al., 2018). The combined increase in hydrogen bonding and disulfide interactions, along with preserved hydrophobic interactions, suggests that TM formulations achieved a well-balanced interaction profile conducive to network formation. Table 4Intermolecular forces and thermal stability of high-moisture meat analogues (HMMAs) containing edible insect proteins.TraitsCON1)TM4TM8PB4PB8GB4GB8Hydrogen bond (mg/mL)0.04 ± 0.01d2)0.10 ± 0.00b0.15 ± 0.00a0.08 ± 0.01bc0.11 ± 0.04b0.10 ± 0.00b0.07 ± 0.00cdHydrophobic interaction (mg/mL)4.44 ± 0.09a4.37 ± 0.04a4.35 ± 0.01a4.12 ± 0.02b3.92 ± 0.06c4.01 ± 0.09c3.96 ± 0.06cDisulfide bond (mg/mL)11.14 ± 0.06c11.97 ± 0.03a11.86 ± 0.64ab11.14 ± 0.26c11.33 ± 0.32bc11.45 ± 0.33abc11.79 ± 0.20abPeak temperature (°C)92.83 ± 0.0994.55 ± 2.0990.67 ± 0.2693.29 ± 2.7393.17 ± 1.6590.58 ± 2.5393.34 ± 0.67Enthalpy (J/g)0.21 ± 0.03a0.22 ± 0.05a0.19 ± 0.02ab0.18 ± 0.01ab0.21 ± 0.02a0.14 ± 0.01b0.14 ± 0.03b1)CON, HMMA without insect proteins; TM4, HMMA with defatted Tenebrio molitor larvae 4%; TM8, HMMA with defatted Tenebrio molitor larvae 8%; PB4, HMMA with defatted Protaetia brevitarsis larvae 4%; PB8, HMMA with defatted Protaetia brevitarsis larvae 8%; GB4, HMMA with defatted Gryllus bimaculatus 4%; GB8, HMMA with defatted Gryllus bimaculatus 8%.2)Different superscript letters (a-d) in a row indicate significant differences between HMMAs (p < 0.05), whereas rows without superscript letters indicate no significant differences among treatments (p > 0.05). Intermolecular forces and thermal stability of high-moisture meat analogues (HMMAs) containing edible insect proteins. CON, HMMA without insect proteins; TM4, HMMA with defatted Tenebrio molitor larvae 4%; TM8, HMMA with defatted Tenebrio molitor larvae 8%; PB4, HMMA with defatted Protaetia brevitarsis larvae 4%; PB8, HMMA with defatted Protaetia brevitarsis larvae 8%; GB4, HMMA with defatted Gryllus bimaculatus 4%; GB8, HMMA with defatted Gryllus bimaculatus 8%. Different superscript letters (a-d) in a row indicate significant differences between HMMAs (p < 0.05), whereas rows without superscript letters indicate no significant differences among treatments (p > 0.05). DSC analysis further supported these findings. While peak temperatures showed minor variations, enthalpy values differed significantly, with GB treatments showing the lowest values (0.14 J/g). Since the extrusion of HMMA generally leads to extensive protein denaturation, further structure formation during the subsequent heating process is limited (Guyony et al., 2023). Thus, the thermal energy supplied during DSC analysis may partially disrupt the established network. The low enthalpy of GB during thermal denaturation indicates reduced structural organization or fewer thermally responsive domains, consistent with weaker interaction networks. Therefore, these results demonstrate that G. bimaculatus may hinder optimal intermolecular interactions required for fibrous structure formation. Microstructure The microstructural characteristics of HMMAs formulated with defatted edible insect proteins are shown in Fig. 1. Microstructural analysis revealed clear differences in fiber formation among treatments. The control sample exhibited a typical anisotropic structure with visible pores, indicating incomplete network consolidation. In contrast, TM4 and TM8 showed more continuous and aligned fibrous structures, with TM4 displaying a more compact and less porous morphology. This dense and well-aligned structure is characteristic of high-quality meat analogues and is associated with desirable mechanical properties and sensory attributes (Dekkers et al., 2018). In HMMAs, because cross-linking through disulfide bonds increases structural stability and decreases the number and size of air cells (Samard et al., 2019), TM4 and TM8, which showed increased disulfide bond formation, formed highly organized structures. PB samples exhibited slight porosity, indicating reduced compactness. This may also be associated with chitin and cuticular proteins as mentioned above, which can disrupt the formation of a continuous protein network. GB treatments also resulted in less organized and porous structures, reflecting weaker matrix formation. Given that GB already possesses more developed muscle proteins with inherent disulfide bonds compared to TM and PB (Kim et al., 2022), these proteins may tend to form partial aggregates rather than fully participating in the formation of an organized structure within the soy-gluten network during extrusion. Overall, the microstructural results confirm that insect species significantly influence HMMA structuring, with T. molitor exhibiting the most favorable performance. Microstructure of high-moisture meat analogues (HMMAs) containing edible insect proteins. Microstructures of HMMAs observed by scanning electron microscopy at 100× and 500× magnifications: (a, b) CON; (c, d) TM4; (e, f) TM8; (g, h) PB4; (i, j) PB8; (k,l) GB4; (m, n) GB8. CON, HMMA without insect proteins; TM4, HMMA with defatted Tenebrio molitor larvae 4%; TM8, HMMA with defatted Tenebrio molitor larvae 8%; PB4, HMMA with defatted Protaetia brevitarsis larvae 4%; PB8, HMMA with defatted Protaetia brevitarsis larvae 8%; GB4, HMMA with defatted Gryllus bimaculatus 4%; GB8, HMMA with defatted Gryllus bimaculatus 8%. Fourier transform infrared spectroscopy (FTIR) The FTIR spectra of HMMAs containing defatted edible insect proteins are shown in Fig. 2. FTIR analysis revealed that insect incorporation induced molecular-level changes in the HMMA matrix (Fig. 2A). All samples exhibited characteristic protein and carbohydrate bands; however, differences were observed in the O–H/N–H stretching (3600–3000 cm−1) and amide regions (1700–1500 cm−1). Blue shifts, especially in TM, and peak broadening, especially by PB, in the O–H/N–H region were observed following insect protein addition, resulting from the increased intermolecular interactions (Zhao et al., 2025). A blue shift of this peak generally is attributed to the decreased hydrogen bonding, and broadening is interpreted as an increase in variety of hydrogen bonds within the matrix. Although the hydrogen bonds in TM-added HMMA did not decrease, the increased disulfide bonds may have affected the O—H stretching. These findings confirm that T. molitor and P. brevitarsis proteins participate in structural alteration and reorganization, contributing to improved network formation. The peaks in the amide I region (1700–1600 cm−1) induced by C Created by potrace 1.16, written by Peter Selinger 2001-2019 O stretching in proteins, were changed, suggesting secondary structural rearrangements of proteins during extrusion (Hou et al., 2023). Changes in C—O stretching and C-O-C vibration (1023–1019 cm−1) can be related to starch–protein interactions, which are important for multiphase structuring in extruded systems. FTIR spectra (a) and secondary structure (b) of high-moisture meat analogues (HMMAs) containing edible insect proteins. CON, HMMA without insect proteins; TM4, HMMA with defatted Tenebrio molitor larvae 4%; TM8, HMMA with defatted Tenebrio molitor larvae 8%; PB4, HMMA with defatted Protaetia brevitarsis larvae 4%; PB8, HMMA with defatted Protaetia brevitarsis larvae 8%; GB4, HMMA with defatted Gryllus bimaculatus 4%; GB8, HMMA with defatted Gryllus bimaculatus 8%. a-cDifferent superscript letters above the bars indicate significant differences between HMMAs (p < 0.05), whereas bars without superscript letters indicate no significant differences among treatments (p > 0.05). Using the changes in amide I region (1700–1600 cm−1), secondary structures of HMMAs were analyzed (Fig. 2B). The α-helix content showed slight decrease in PB4 and the lowest content in GB8, while β-sheet content did not differ significantly among treatments (p > 0.05). The β-turn content increased with the addition of TM and PB, whereas the random coil content was highest in GB8. α-helices and β-sheets are more ordered and dense structures compared to β-turns and random coils, and the changes in their ratio suggest that insect protein addition affected the intramolecular interaction within proteins (Guo et al., 2020). Thus, the decreased α-helix content and increased random coil content observed in GB8 imply increased structural disorder and decreased conformational stability at the higher inclusion level of G. bimaculatus (Choi et al., 2024). Meanwhile, since the increase in β-turn content relates to the changes in polypeptide chain directions, protein folding and interactions may be facilitated by TM and PB addition (Huang et al., 2024). The increase in β-turn content, accompanied by a slight decrease in α-helix content (p > 0.05), may indicate a shift toward a less ordered protein conformation, which appears inconsistent with the microstructural observation (Bondu, Gimeno, Evon, Vaca-Medina and Rouilly, 2024). Nevertheless, a previous study has reported that increases in β-turn induced by the incorporation of protein ingredient were significantly and positively correlated with the degree of fibrous structuring (Tian et al., 2023). Thus, the increase in β-turn content of TM-added HMMAs may affect fibrous structure with increased intermolecular interactions. Therefore, these results suggest that the effects of insect protein incorporation on the intramolecular interactions and secondary structure within HMMAs were species-dependent. Texture profile analysis The texture profile parameters of HMMAs formulated with defatted edible insect proteins are presented in Table 5. Texture is a critical quality attribute of high-moisture meat analogues because it directly determines consumer perception of meat-like bite, chewiness, structural integrity, and fibrous tearing behavior. The textural properties were significantly influenced by insect incorporation, reflecting structural differences at both the molecular and microstructural levels. Hardness varied notably among treatments, with TM4 exhibiting the highest value, followed by TM8, indicating that moderate inclusion (4%) resulted in the most mechanically resistant structure, whereas the other treatments showed significantly lower values. This trend is consistent with the microstructural observations, where TM displayed the most compact and continuous fibrous network. In high-moisture meat analogues, hardness is closely associated with matrix density and protein–protein interaction strength, indicating that T. molitor contributes to a mechanically robust structure. Springiness showed moderate variation but were generally higher in TM and PB samples than in GB samples. TM8, PB4, and PB8 exhibited the highest springiness, suggesting improved elastic recovery after deformation. Cohesiveness showed the highest value at CON, followed by TM4 and PB4, indicating better internal structural integrity during compression. These results suggest that insect proteins, particularly those of T. molitor and P. brevitarsis, enhance the continuity and resilience of the protein network, which are critical for a desirable meat-like texture. Gumminess and chewiness, derived from hardness, cohesiveness, and springiness, were significantly higher in TM formulations, especially TM4. This indicates that TM-based HMMAs require greater energy for mastication, closely resembling conventional meat texture. PB samples exhibited intermediate values, whereas GB samples showed the lowest values, consistent with their more porous and less organized microstructure. The texturization index, defined as the ratio of shear force in the longitudinal to transverse direction, further highlights differences in anisotropic structure formation. TM4 exhibited the highest value, indicating the strongest directional fibrous structure. In contrast, PB4 and GB4 showed significantly lower values, reflecting weaker fiber alignment and reduced anisotropy. The development of anisotropic structure is a key quality attribute in HMMAs, as it directly influences bite and tear behavior (Dekkers et al., 2018). Table 5Texture profiles of high-moisture meat analogues (HMMAs) containing edible insect proteins.TraitsCON1)TM4TM8PB4PB8GB4GB8Hardness (N)203.67 ± 15.10cd2)286.74 ± 12.33a236.69 ± 13.56b221.22 ± 14.41bc216.48 ± 16.81cd198.10 ± 15.95d221.93 ± 13.05bcSpringiness0.91 ± 0.080.90 ± 0.020.93 ± 0.010.93 ± 0.040.93 ± 0.050.85 ± 0.040.90 ± 0.02Cohesiveness0.90 ± 0.01a0.86 ± 0.02b0.84 ± 0.00c0.86 ± 0.03b0.85 ± 0.01bc0.85 ± 0.01bc0.84 ± 0.01bcGumminess (N)182.52 ± 13.90bc248.68 ± 13.54a193.97 ± 7.30b189.88 ± 4.13bc188.24 ± 15.68bc177.02 ± 12.84c188.53 ± 10.25bcChewiness (N)159.70 ± 19.82c222.65 ± 10.68a186.15 ± 8.90b175.22 ± 5.68bc184.84 ± 8.41b143.02 ± 10.80d167.73 ± 5.73cTexturization index2.86 ± 0.06b3.47 ± 0.05a2.53 ± 0.10c2.14 ± 0.05f2.40 ± 0.02d2.27 ± 0.04e2.86 ± 0.10b1)CON, HMMA without insect proteins; TM4, HMMA with defatted Tenebrio molitor larvae 4%; TM8, HMMA with defatted Tenebrio molitor larvae 8%; PB4, HMMA with defatted Protaetia brevitarsis larvae 4%; PB8, HMMA with defatted Protaetia brevitarsis larvae 8%; GB4, HMMA with defatted Gryllus bimaculatus 4%; GB8, HMMA with defatted Gryllus bimaculatus 8%.2)Different superscript letters (a-f) in a row indicate significant differences between HMMAs (p < 0.05), whereas rows without superscript letters indicate no significant differences among treatments (p > 0.05). Texture profiles of high-moisture meat analogues (HMMAs) containing edible insect proteins. CON, HMMA without insect proteins; TM4, HMMA with defatted Tenebrio molitor larvae 4%; TM8, HMMA with defatted Tenebrio molitor larvae 8%; PB4, HMMA with defatted Protaetia brevitarsis larvae 4%; PB8, HMMA with defatted Protaetia brevitarsis larvae 8%; GB4, HMMA with defatted Gryllus bimaculatus 4%; GB8, HMMA with defatted Gryllus bimaculatus 8%. Different superscript letters (a-f) in a row indicate significant differences between HMMAs (p < 0.05), whereas rows without superscript letters indicate no significant differences among treatments (p > 0.05). These textural differences are directly linked to the intermolecular interaction patterns discussed in Section 3.4. The superior performance of TM formulations can be attributed to a balanced combination of hydrophobic interactions, enhanced hydrogen bonding, and increased disulfide cross-linking, which collectively promote protein alignment and network stabilization during extrusion (Schmid, Farahnaky, Adhikari and Torley, 2022, Zhang et al., 2023). In contrast, reduced hydrophobic interactions and lower thermal stability in GB samples likely limited effective network formation, resulting in weaker mechanical properties. Furthermore, the improved textural properties of TM samples may also be partially explained by their amino acid composition, particularly the higher levels of sulfur-containing amino acids and flexible residues such as glycine and alanine, which facilitate protein rearrangement and cross-linking during processing. Similar relationships among protein composition, intermolecular interactions, and texture development have been reported in extruded protein systems (Schmid, Farahnaky, Adhikari and Torley, 2022, Zhang et al., 2023). Overall, texture profile analysis confirms that edible insect proteins significantly modulate the mechanical properties of HMMAs in a species-dependent manner. Although TM8 showed a compact and well-aligned microstructure, the highest instrumental texturization metrics (hardness, chewiness, and texturization index) were observed for TM4, suggesting that excessive compaction at 8% inclusion does not necessarily translate into maximal directional anisotropy. PB formulations also showed promising performance at moderate inclusion levels, whereas GB was less effective as a structuring agent under these conditions. Conclusion This study demonstrated that the incorporation of Tenebrio molitor, Protaetia brevitarsis, and Gryllus bimaculatus significantly affected the physicochemical, nutritional, structural, and textural properties of soy–gluten-based HMMAs in a species-dependent manner. Furthermore, the effects of inclusion levels also varied in a species-dependent manner, modifying matrix development during high-moisture extrusion. Overall, the addition of edible insects significantly enhanced the nutritional quality of the products due to the enrichment of essential amino acids. Regarding intermolecular interactions, disulfide bonds, which are the main interactions in HMMAs, were increased by insect protein incorporation, with TM showing the highest degree. Due to the increased disulfide bond formation, a fibrous and dense structure was formed, resulting in the highest textural properties with 4% TM. However, 8% TM and other insect protein additions induced a less anisotropic structure with more air cells. These results suggest that their specific protein compositions and the incorporation of cuticular fractions may have impeded integration into the soy-gluten-based matrix, thereby obstructing the formation of a stable and dense fibrous structure. Consequently, although edible insects can function as additives for nutritional enrichment in HMMAs, Tenebrio molitor larvae addition in a moderate amount can be a suitable choice for improving physicochemical properties and mimicking a meat-like structure. Meanwhile, future research should comprehensively address sensory evaluation, including flavor profiling and consumer acceptability tests. In addition, allergenicity and toxicological assessments using appropriate in vitro and cell-based approaches should be conducted to ensure food safety before widespread practical application of novel insect-based foods. CRediT authorship contribution statement Yea-Ji Kim: Writing – review & editing, Writing – original draft, Methodology, Formal analysis, Conceptualization. Ji Yoon Cha: Writing – review & editing, Writing – original draft, Methodology, Formal analysis. Jeong-Heon Kim: Writing – review & editing, Writing – original draft, Methodology, Formal analysis. Dong Hyun Keum: Writing – review & editing, Writing – original draft, Methodology, Formal analysis. Bon-Jae Gu: Writing – review & editing, Writing – original draft, Methodology, Data curation. Yun-Sang Choi: Writing – review & editing, Writing – original draft, Supervision, Methodology, Investigation, Formal analysis, Conceptualization. Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Acknowledgements This study was supported by the Main Research Program (E211200-06) of the Korea Food Research Institute (KFRI), funded by the Ministry of Science and ICT, Republic of Korea. This research was also supported by the Rural Development Administration (RDA), Republic of Korea, through the Cooperative Research Program for Agriculture Science & Technology Development (Project No. RS-2026-25508714). Footnotes Supplementary data to this article can be found online at https://doi.org/10.1016/j.fochx.2026.104345. Supplementary data Supplementary material Data availability No data was used for the research described in the article. References Associated Data Supplementary Materials Supplementary material Data Availability Statement No data was used for the research described in the article.