Bioactive camel skin gelatin hydrolysates: Functional stability and application in edible coatings for cherry tomato preservation

TL;DR

Researchers made gelatin from camel skin that can extend the shelf life of cherry tomatoes by reducing spoilage and microbes.

Contribution

Camel skin gelatin hydrolysates (CSGH_S50) show stable antioxidant and antimicrobial properties suitable for edible coatings.

Findings

CSGH_S50 showed strong antioxidant activity and stability under various conditions.

CSGH_S50 extended cherry tomato shelf life by reducing microbial growth and spoilage.

Savinase produced higher hydrolysis degree gelatin than Neutrase.

Abstract

This study aimed on the production of camel skin gelatin hydrolysates (CSGHs) using Savinase and Neutrase at varying hydrolysis levels. The Savinase-derived fraction (CSGH_S50) showed potent antioxidant capacity in terms of DPPH• radical-scavenging, ferric reducing power, β-carotene-linoleate bleaching and ferrous-ion chelating activities. Interestingly, this fraction maintained its activity under wide pH variations, prolonged thermal treatment and simulated gastrointestinal digestion. In parallel, the Neutrase-derived fraction (CSGH_N50) showed notable antimicrobial activity against Salmonella enterica. Cherry tomatoes, known for their delicate skin and high respiration rate, were used as a model of highly perishable fresh produce. Coating experiments revealed that incorporating CSGH_S50 into camel skin gelatin significantly enhanced key quality attributes of cherry tomatoes during storage. The enriched coatings effectively reduced deterioration and microbial growth, thereby extending the fruits' shelf-life. Overall, these findings demonstrated the potential of CSGHs as natural, multifunctional agents for improving the preservation and safety of fresh products.

•Savinase produced CSGHs with a higher hydrolysis degree than Neutrase.•CSGH_S50 showed superior in vitro antioxidant and antimicrobial activities.•CSGH_S50 remained stable under pH, thermal, and simulated digestive conditions.•CSGH_S50 coating extended tomato shelf-life by reducing spoilage and microbes.

Savinase produced CSGHs with a higher hydrolysis degree than Neutrase.

CSGH_S50 showed superior in vitro antioxidant and antimicrobial activities.

CSGH_S50 remained stable under pH, thermal, and simulated digestive conditions.

CSGH_S50 coating extended tomato shelf-life by reducing spoilage and microbes.

Figures

Degree of hydrolysis (%) as a function of hydrolysis time. Hydrolysates obtained with Neutrase at 30 and 50 U per mg of substrate are labeled CSGH_N30 and CSGH_N50, respectively, while those produced with Savinase at the same ratios are labeled CSGH_S30 and CSGH_S50. *(U/mg refers to units per mg of substrate, not specific enzyme activity.)*Fig. 1

Antioxidant activities of CSGHs measured by (A) DPPH• scavenging, (B) ferric reducing power, (C) Metal chelating, and (D) β-carotene bleaching assays.Fig. 2

Antioxidant stability of CSGH_S50. (A) pH-dependent residual activities, calculated relative to the activity at pH 7 (values >100% indicate higher activity than the reference). (B) Thermal stability: residual activities *vs* heating time, relative to the untreated control (0 min). (C) Effect of *in vitro* gastrointestinal digestion: residual activities relative to the undigested control (0 min). Assays in all panels include β-carotene-linoleate bleaching, ferric reducing power, and DPPH• radical-scavenging activity.Fig. 3

Changes in cherry tomato quality attributes during storage: (A) weight loss, (B) total color difference (ΔE), (C) pH, (D) total titratable acidity (TTA) and (E) firmness. Treatments were: C, control (uncoated); G1, 3% glycerol; G2, 3% glycerol +2% CSGH_S50; G3, 3% CSG + 2% CSGH_S50.Fig. 4

Cherry tomato appearance during storage. Treatments were: C, control (uncoated); G1, 3% glycerol; G2, 3% glycerol + 2% CSGH_S50; G3, 3% CSG + 2% CSGH_S50.Fig. 5

Full text

Introduction

1

The preservation of fresh products, especially highly perishable fruits such as cherry tomatoes, represents a critical challenge for the food industry, which must maintain safety, nutritional quality, and sensory attributes during storage and distribution. Postharvest losses arise primarily from water loss, microbial contamination, and biochemical degradation, all of which compromise quality and reduce shelf-life (Zaman et al., 2025). Globally, postharvest food loss is estimated to be around 30% of total production (Lalpekhlua et al., 2024), largely due to suboptimal post-harvest management practices, inadequate storage conditions, and inefficient handling processes. In the case of tomatoes, losses can be as high as 42% globally, depending on region and handling conditions (Arah et al., 2015). These losses can be both quantitative (weight) and qualitative (nutritional or visual quality) impacts, which negatively affect the profitability of tomato production, particularly in developing countries (Molelekoa et al., 2025). While synthetic coatings and chemical preservatives remained effective in delaying spoilage, rising consumer demand for natural, clean-label solutions had stimulated intensive research into bioactive, renewable packaging materials (Sheibani et al., 2024).

The future of edible coatings depends on the design of advanced biopolymeric matrices that combine barrier, mechanical, and bioactive functionalities. Among natural biopolymers, gelatin-based coatings have attracted considerable attention for their ability to form protective films that minimize moisture loss, inhibit microbial proliferation, and delay oxidative reactions in fruits and vegetables. Compared to conventional synthetic coatings, edible coatings offered several advantages, including biodegradability, non-toxicity, and the ability to incorporate natural bioactive compounds**,** making them safer and more environmentally sustainable (Mezhoudi et al., 2022; Sayah et al., 2024). Furthermore, the incorporation of protein hydrolysates into gelatin matrices can significantly enhance their preservative performance by introducing intrinsic antioxidant and antimicrobial activities, thereby improving the overall shelf-life, safety, and quality of fresh produce (Mohammadi-Moghaddam et al., 2023).

Gelatin is one of the most valuable proteins in the food industry, and it can be deried from different by-product sources (bovine, procune, marine, etc.). The search of underutilized sources, such as camel by-products, has gained growing interest. Camel skin gelatin (CSG) is a promising candidate due to its unique amino acid (AA) composition and favorable techno-functional characteristics, including high gel strength, excellent film-forming ability, and complete solubility in water. Its structural features also confer interesting emulsifying and foaming properties, making it suitable for the development of biodegradable edible films and coatings. Moreover, valorizing camel skin contributes to sustainable resource management and enhances the economic value of the camel meat industry (Hajlaoui et al., 2024).

Enzymatic hydrolysis represents an efficient and environmentally friendly approach to enhance the functional and biological value of proteins. This process avoids the use of agressive solvents and chemicals, while generating short-chain peptides with potential bioactivities. Its efficiency depends largely on the specificity and catalytic behavior of the protease employed, as different enzymes cleave peptide bonds at distinct sites, thereby influencing the degree of hydrolysis (DH), peptide profile, and bioactivity. Savinase and Neutrase are two industrially relevant proteases with contrasting cleavage mechanisms and substrate specificities. However, their comparative efficiency in releasing bioactive peptides from camel skin gelatin has not yet been investigated. Optimizing hydrolysis parameters, particularly the enzyme-to-substrate (E/S) ratio, is also important for maximizing peptide yield and biological activity. A higher DH generally correlates with increased exposure of AAs such as Met, His, and Phe residues known for their antioxidant roles in radical scavenging and metal ion chelation (Borrajo et al., 2020). Such peptides offer a natural strategy to mitigate oxidative damage and microbial spoilage in food systems.

The present study aimed to produce and characterize gelatin hydrolysates from camel skin (CSGHs) as part of a novel valorization approach for this underutilized by-product. Their antioxidant properties were evaluated using complementary assays, including DPPH**•** radical-scavenging, ferric reducing power, β-carotene-linoleate bleaching, and ferrous-ion chelating activities. The antimicrobial activity was assessed against representative Gram-positive, Gram-negative, and fungal foodborne pathogens. Finally, the most active hydrolysate was incorporated into a gelatin-based edible coating to investigate its practical application as a bioactive ingredient for extending the shelf-life of cherry tomatoes.

Materials and methods

2

Chemicals and biological materials

2.1

Dromedary camel (Camelus dromedarius) skin was obtained from a slaughterhouse in Sfax, Tunisia. All chemicals, enzymes and microbiological media (Luria-Bertani (LB) agar and peptone Potato Dextrose Agar (PDA)) were from Sigma-Aldrich (St. Louis, MO, USA). Enzymes and reagents used are the following: Pepsin (EC 3.4.23.1, CAS 9001-75-6), Neutrase® (EC 3.4.24.28, CAS 9014-01-1), Savinase® (EC 3.4.21.62, CAS 9014-01-1), Pancreatin (CAS 8049-47-6, mixture of protease, amylase, and lipase activities), Trypsin (EC 3.4.21.4, CAS 9002-07-7), bile extract (sodium deoxycholate, CAS 232–369-0), glycerol (CAS 56–81-5), Ca(OH)2 (CAS 1305-62-0), (NH_4_)2_SO_4 (CAS 7783-20-2), NaOH (CAS 1310-73-2), phenyl isothiocyanate (PITC) (CAS 103–72-0), acetonitrile (CAS 75–05-8), 2,2-diphenyl-1-picrylhydrazyl (DPPH**•**, CAS 1898-66-4), trichloroacetic acid (CAS 76–03-9), FeCl_2_ 4H_2_O (CAS 13478–10-9), 3-(2-pyridyl)-5,6-diphenyl-1, acetonitrile 2,4-triazine-p,p’-disulfonic acid monosodium salt hydrate (ferrozine, CAS 63451–29-6), β-carotene (CAS 7235-40-7), linoleic acid (CAS 60–33-3), Tween-40 (CAS 9005-66-7), KCl (CAS 7447-40-7), NaHCO_3_ (CAS 144–55-8), phenolphthalein (CAS 68807–90-9), ethanol (CAS 64–17-5), methanol (CAS 67–56-1), K_3_[Fe(CN)6] (CAS 13746–66-2), FeCl_3_ (CAS 7705-08-0), chloroform (CAS 67–66-3) and HCl (CAS 7647-01-0). The phosphate buffer was prepared using Na_2_HPO_4_ (CAS 7558-79-4) and NaH_2_PO_4_ (CAS 7558-80-7).

Camel skin preparation

2.2

Fresh camel skins were placed in sterile polyethylene bags immediately after slaughter and transported to the laboratory under controlled temperature (4 °C), arriving within 30 min to minimize degradation. In the laboratory, skins were thoroughly washed with distilled water and residual muscle fibers and subcutaneous fat were manually removed using surgical scalpels*.* For batch homogeneity, several skin samples were pooled and stored at −20 °C. Before use, the frozen samples were thawed gradually (72 h at 4 °C) and then cut into standardized strips (3.0 × 1.0 cm) using precision cutting tools to ensure uniform treatment surfaces*.*

Protease activity

2.3

The method of Kembhavi et al. (1993), using casein as a substrate, was used to measure the activity of Neutrase and Savinase. One unit of protease activity was defined as the amount of enzyme required to release 1 μg of tyrosine per minute under the experimental conditions.

Gelatin extraction and preparation of protein hydrolysates

2.4

CSG was extracted following to Hajlaoui et al. (2024), with minor adjustments to enzymatic and heat treatments. Approximately 1 kg of fresh camel skin were treated with 10% (w/v) Ca(OH)2 at 25 °C for 24 h with gentle stirring, rinsed with distilled water, and then their pH was adjusted to neutrality using 3% (w/v) (NH_4_)2_SO_4. After pH neutralization, the skin was incubated at 70 °C for 48 h in a water bath equipped with a magnetic stirrer to ensure continuous agitation and facilitate gelatin extraction. Following extraction, gelatin was centrifuged at 10,000 ×g for 30 min to remove insoluble material, and the supernatant was subsequently freeze-dried with a Moduloyd freeze-dryer (Bioblock Scientific Christ ALPHA 1–2, IllKrich-Cedex, France), producing CSG powder. Based on the previous work of Hajlaoui et al. (2024), the extraction yield was 29.16 ± 1.56% (fresh weight basis).

For gelatin hydrolysates, 3 g of CSG (3% w/v, chosen based on preliminary experiments to ensure complete solubilization and optimal enzyme-substrate interaction) were dissolved in 100 mL of 10 mM phosphate or Tris-HCl buffer, then adjusted to the optimal pH and temperature of each protease (Neutrase: 7.0, 50 °C; Savinase: 9.0, 50 °C). After 30 min of equilibration, hydrolysis was initiated by adding the enzyme stock solution (expressed in U/mL) to achieve enzyme/substrate ratios of 30 or 50 U of enzyme per mg of gelatin. These ratios were selected based on preliminary optimization experiments (data not shown) to ensure sufficient hydrolysis within a practical time frame while avoiding excessive degradation of gelatin peptides. The pH was maintained at its optimum by automated addition of 4 N NaOH. Following thermal inactivation at 90 °C for 10 min, the mixture was centrifuged at 5000 ×g for 30 min, and the resulting supernatants were freeze-dried and stored at −20 °C.

The DH, which indicates the fraction of peptide bonds hydrolyzed in the substrate, was determined according to Jridi et al. (2014) for all samples. Hydrolysates obtained with Neutrase at 30 U/mg and 50 U/mg were referred to as CSGH_N30 and CSGH_N50, respectively, whereas those produced with Savinase at the same ratios were designated as CSGH_S30 and CSGH_S50, respectively.

Chemical analysis

2.5

The moisture, ash, protein, and fat contents of CSG and CSGH powders were determined according to the methods described by the Association of Official Analytical Chemists (AOAC, 2000). The protein content was determined using the Kjeldahl method with a nitrogen-to-protein conversion factor of 5.5. All measurements were performed in triplicate.

Amino acid composition

2.6

The AA composition of CSGHs was determined by complete acid hydrolysis, followed by separation and quantification using an HPLC system equipped with a PicoTag® column (300 mm × 3.9 mm, Waters, Milford, MA, USA) coupled to diode array detector. Briefly, 50 μL aliquots of CSGH solutions (3 mg/mL) were hydrolysed in 300 μL of 6 N HCl containing 1% phenol at 120 °C for 24 h. Hydrolysates were derivatized with PITC and dissolved in 300 μL of 5 mM sodium phosphate buffer (pH 7.4) containing 5% acetonitrile.

The analysis employed a binary gradient system consisting of (A) 0.07 M sodium acetate buffer (pH 6.55) with 2.5% acetonitrile and (B) acetonitrile/water/methanol mixture (45,40,15, v/v/v). The temperature was maintained at 52 °C with detection at 254 nm. A multi-step gradient elution was applied, starting with 3% B at 13.5 min, increasing to 40% B by 50 min, followed by column washing with 100% B for 10 min. Amino acid contents were quantified based on peak areas and expressed as relative percentages of the total composition, corresponding to residues per 100 residues (total AA composition).

Antimicrobial activity

2.7

Microbial strains

2.7.1

Antibacterial activity of CSGHs was evaluated against a panel of ten bacterial strains, including five Gram-positive species: Bacillus cereus (ATCC 11778), Listeria monocytogenes (ATCC 19117), Micrococcus luteus (ATCC 4698), Staphylococcus aureus (ATCC 25923), and Enterococcus faecium (ATCC 29212), and five Gram-negative species: Escherichia coli (ATCC 25922), Klebsiella pneumoniae (ATCC 13883), Pseudomonas fluorescens (ATCC 13525), Salmonella enterica (ATCC 10708), and Salmonella typhi (ATCC 19430).

Antifungal activities were tested using Botrytis cinerea and Fusarium oxysporum. The microorganisms were obtained from the culture collection of the Laboratory of Functional Physiology and Valorization of Bio-resources (LR23ES08) at ISBB, Beja, Tunisia.

Agar diffusion method

2.7.2

The antimicrobial activity of CSGHs was evaluated using the agar diffusion method, as previously described (Vanden Berghe & Vlietinck, 1991) with minor modifications.

For antibacterial activity, bacterial inocula were prepared by adjusting overnight cultures to 10^6^ colony-forming units (cfu)/mL in sterile broth. Indicator bacterial cultures were spread onto LB agar plates, and 50 μL of each sample (20 mg/mL in distilled water) was placed into 6 mm wells made in the LB agar. The plates were stored at 4 °C for 1 h to allow diffusion, then incubated at 37 °C for 24 h.

For antifungal activity, fungal spore suspensions (10^8^ spores/mL) were spread onto PDA plates, and 50 μL of each sample (20 mg/mL in distilled water) was placed into 6 mm wells made in the agar. Plates were stored in the dark at 4 °C for 2 h to allow diffusion of the samples, then incubated at 30 °C for 72 h.

For both bacterial and fungal assays, antimicrobial activity was determined by measuring the diameter (mm) of the inhibition zones surrounding the wells. All tests were performed in triplicate, and results were averaged.

Antioxidant properties

2.8

DPPH• radical-scavenging activity

2.8.1

The DPPH**•** radical-scavenging activity was assessed following the method described by Bersuder et al. (1998). Test solutions (500 μL) at concentrations ranging from 0 to 5 mg/mL, or deionized water as a negative control, were combined with 325 μL of ethanol and 125 μL of DPPH**•** solution (0.2 mM in ethanol). Mixtures were incubated in the dark at 25 °C for 60 min, and absorbance was recorded at 517 nm using UVmc® spectrophotometer (SAFAS, Monaco). The selected concentration range (0–5 mg/mL) was chosen based on preliminary tests to ensure measurable and comparable radical-scavenging activity without reaching saturation. The scavenging activity was determined as:

$$[eqn]$$

where AC and AS are the absorbance of control and sample, respectively. Each test was performed in triplicate, and values were reported as means.

Ferric reducing power

2.8.2

The ferric reducing power of the samples was assessed by the spectrophotometric method of Benzie and Strain (1996). Test solutions (0.5 mL) at a concentrations ranging from 0 to 5 mg/mL were mixed with 0.2 M phosphate buffer (pH 6.6) and 1% K_3_[Fe(CN)6], then incubated at 50 °C for 20 min. Trichloroacetic acid (10%) was added, and the mixture was centrifuged at 3000 ×g for 10 min. The resulting supernatant was then combined with FeCl_3_ solution for reaction. The formation of ferrous-ferricyanide complex was measured at 700 nm using UVmc® spectrophotometer (SAFAS), where increased absorbance correlated with enhanced reducing capacity.

Ferrous-ion chelating activity

2.8.3

The metal chelating activity of CSGHs was determined using the method of Decker and Welch (1990) with minor adjustments. Briefly, 50 μL of 2 mM FeCl_2_ 4H_2_O was added to 100 μL of the sample (0–5 mg/mL) diluted with 450 μL of water and incubated at ambient temperature for 5 min. The reaction was initiated by adding 200 μL of 5 mM ferrozine solution, followed by thorough mixing and a 10 min incubation. The absorbance was then recorded at 562 nm using UVmc® spectrophotometer (SAFAS). Water was used as the negative control. The chelating activity was expressed as:

$$[eqn]$$

where AC and AS correspond to the absorbance of control and sample, respectively.

β-Carotene-linoleate bleaching activity

2.8.4

The antioxidant activity of CSGHs was assessed using the β-carotene-linoleate model system according to Koleva et al. (2002). A stock emulsion was prepared by dissolving 0.5 mg of β-carotene in 1 mL chloroform containing 25 μL linoleic acid and 200 μL Tween-40. After evaporating the solvent at 40 °C under vacuum, 100 mL of distilled water was added with vigorous stirring to obtain the working emulsion. For the assay, 2.5 mL of this emulsion was mixed with 0.5 mL of CSGH solution (0–5 mg/mL) and incubated at 50 °C for 2 h. Absorbance measurements at 470 nm, using UVmc® spectrophotometer (SAFAS), were taken against a blank (emulsion without β-carotene) both initially (A_0_) and after incubation (A_120_). A negative control (water instead of sample) was used. Antioxidant activity was calculated as percentage inhibition of β-carotene-linoleate bleaching:

$$[eqn]$$

where subscripts s and c represent sample and control measurements, respectively.

Stability of gelatin hydrolysate (CSGH_S50)

2.9

pH stability

2.9.1

The pH stability was determined as described by Jridi et al. (2014). CSGH_S50 was dissolved in 10 mL of distilled water to obtain a final concentration of 50 mg/mL (protein basis). Aliquots were adjusted to pH values ranging from 1.0 to 11.0 using 1 M HCl or 1 M NaOH, then the volume was brought to 25 mL with distilled water. The pH value was measured using PHS-3E pH meter (Shanghai Leici Scientific Instrument Co., Shanghai, China). Samples were incubated at 25 ± 2 °C for 1 h. After incubation, the pH of all samples was readjusted to pH 7.0 using the same titrants, and the final volume was made up to 50 mL with distilled water. A control sample was prepared by dissolving CSGH_S50 to the same initial concentration and adjusting the pH to 7.0 before incubation, then subjected to the same dilution, incubation, and final volume adjustments as the treated samples.

The residual antioxidant activities of pH-treated solutions were evaluated using the β-carotene-linoleate bleaching, DPPH• radical-scavenging and ferric reducing power assays, following the procedures described earlier. Results were expressed as residual activity (%), calculated as follows:

$$[eqn]$$

All measurements were performed in triplicate.

Thermal stability

2.9.2

Thermal stability was determined as described by Jridi et al. (2014) with slight modifications. CSGH_S50 was dissolved in 10 mL of distilled water at a protein concentration of 50 mg/mL. The pH of the solution was adjusted to 7.0, and the volume was brought to 50 mL with distilled water. Aliquots of 10 mL were transferred into screw-capped test tubes and placed in a boiling water bath (100 °C) for 0, 15, 30, 60, 120, 180, and 240 min. Then, the tubes were immediately cooled in iced water. The selection of thermal stability time points was guided by preliminary experiments, which showed a rapid change in activity within the first hour of heating, followed by a slower decline during prolonged exposure. This time points allowed us to capture both the initial inactivation phase (15–60 min) and the extended thermal behavior of the hydrolysate (120–240 min), providing a comprehensive view of its stability profile.

A control sample was prepared by dissolving CSGH_S50 at the same concentration, adjusting the pH to 7.0, and keeping it at room temperature without thermal treatment; it was subjected to the same handling steps (transfer to tubes, cooling, dilution, and storage) to ensure procedural equivalence. The residual antioxidant activities of heat-treated samples were determined using the the β-carotene-linoleate bleaching, DPPH• radical-scavenging and ferric reducing power methods, following the procedures described earlier.

Results were presented as residual activity of the sample (at different time points of heat) expressed as a percentage of the initial activity (0 min, untreated control), using the following formula.

$$[eqn]$$

All measurements were performed in triplicate.

In vitro gastrointestinal digestion

2.9.3

The effect of simulated gastrointestinal digestion on CSGH_S50 was evaluated following the procedure of Jridi et al. (2014) with slight modifications. Briefly, 100 mL of CSGH_S50 solution (10 mg/mL) was mixed with 10 mL of phosphate buffer (10 mM, pH 6.8) and pre-incubated at 37 °C for 2 min.

Gastric digestion was initiated by adding 0.5 mL of HCl-KCl buffer (1 M, pH 1.5) to acidify the mixture, followed by the addition of pepsin (32 U/mL, prepared in 1 M HCl-KCl buffer, pH 1.5). The mixture was then incubated at 37 °C for 60 min. Although gastric digestion is often simulated for 2 h, a 1-h duration was selected based on preliminary optimizations indicating that pepsin activity and pH stability were maintained within this time frame under the experimental conditions. After gastric digestion, the pH was adjusted to 6.8 by adding 1 mL of 1 M NaHCO_3_. Duodenal digestion was simulated by adding 1 mL of an enzyme mixture containing pancreatin (10 mg/mL), trypsin (14,600 U/mL), and bile extract (13.5 mg/mL) prepared in 10 mM phosphate buffer (pH 8.2). Samples were incubated at 37 °C for 3 h.

To achieve duodenal digestion, tubes were immersed in boiling water for 10 min to inactivate enzymes, then immediately cooled in ice water. Samples were withdrawn at 0 min (control), 30, 60, 120, 180 and 240 min, during digestion. Antioxidant activities were assessed using the β-carotene-linoleate bleaching, DPPH• radical-scavenging and ferric reducing power methods, following the procedures described earlier. The rationale for evaluating antioxidant activity over time was to monitor potential structural changes and peptide release at different digestion stages, which may enhance or decrease the activity. Results of antioxidant activities were expressed as residual activity (%) relative to the undigested control, using the following formula:

$$[eqn]$$

All measurements were performed in triplicate.

Application – Effect of CSGH-based edible coating on the shelf life of cherry tomatoes

2.10

Edible coatings were prepared following Faisal et al. (2025) and Mezhoudi et al. (2022), with slight modifications. Glycerol, CSG, and its enzymatic hydrolysate (CSGH_S50, at 2% w/v, chosen based on preliminary experiments to ensure sufficient film-forming capacity and bioactive functionality without compromising coating uniformity and adhesion) were dissolved in 100 mL of distilled water and homogenized at 500 rpm for 10 min at room temperature (21 ± 2 °C).

Fresh cherry tomatoes were purchased from the local Béja market (Tunisia) and selected based on their regular size (8 g ± 0.2 g), color (light-red ripening stage (USDA stages 4–5, approximately 60–70% red color development) USDA, 1991), moderate firmness, intact surface, and absence of visible defects or disease. They were fully red and showed no signs of physical damage or fungal assault. All the cherry tomatoes were cleaned and washed to remove dirt. Fresh cherry tomatoes were transported under refrigerated conditions (4 °C; 70% RH) and were used within 24–48 h. Fruits were allowed to equilibrate at room temperature (21 ± 2 °C) for 1 h before coating application.

Fruits were first disinfected by immersion in a sodium hypochlorite solution (200 mg/kg) for 3 min, rinsed with potable water, and air-dried for 30 min. Tomatoes were then randomly distributed into four groups: Control (uncoated); G1: 3% glycerol coatning; G2: 3% glycerol +2% CSGH_S50; and G3: 3% CSG + 2% CSGH_S50. Coatings were applied by dipping each fruit into the respective solution for 2 min, allowing excess liquid to drain. Coated fruits were partially dried in an air-circulating oven at 21 ± 2 °C for 30 min before packaging as previously described by Faisal et al. (2025). Finally, all samples were packed in 64-μm polyethylene pouches and stored at at room temperature. Both coated and control tomatoes were analyzed over 21 days of storage, with sampling performed every 7 days. At each sampling point, ten fruits per treatment were collected for analysis.

Weight loss

2.10.1

Weight loss was determined gravimetrically following the method of AOAC (2000). Five cherry tomatoes from each treatment group, stored in plastic trays, were weighed non-destructively at each sampling point. Weight loss was expressed as the percentage reduction relative to the initial weight of the fruits.

pH

2.10.2

The pH was measured according to the AOAC (2000) method using a PHS-3E pH meter (Shanghai Leici Scientific Instrument Co.). For each treatment, 10 g of cherry tomatoes tissue were homogenized in 100 mL of distilled water, and the pH of the resulting slurry was recorded in triplicate.

Total color difference (ΔE)

2.10.3

The color parameters of the tomato surface were analyzed using a CR-400 colorimeter (Konica Minolta, Tokyo, Japan), as described by Imeneo et al. (2021). The L* (lightness), a* (redness/greenness), and b* (yellowness/blueness) values were recorded. Color measurements obtained on day 0 were used as the reference. The ΔE at each sampling point was calculated using the following formula:

$$[eqn]$$

where $[eqn]$ L*, $[eqn]$ a* and $[eqn]$ b* represent the differences between the color values at a given sampling time and those recorded on day 0. A standard white calibration plate (L_0_ = 97.5, a_0_ = −0.1, b_0_ = 2.3) was used to calibrate the colorimeter before measurements.

Total titratable acidity (TTA)

2.10.4

Five grams of cherry tomato tissue were homogenized in 50 mL of distilled water for 2 min using a Waring blender (Waring Products Co., McConnellsburg, PA). The homogenate was filtered through gauze, and 40 mL of the filtrate were collected for titration (Dong et al., 2018). Two drops of 1% phenolphthalein solution were added as an indicator, and the sample was titrated with 0.1 N NaOH until a faint pink color persisted for at least 30 s. TTA was expressed as grams of citric acid per100 mL using the equation:

$$[eqn]$$

where VNaOH is the volume of 0.1 N sodium hydroxide solution used for titration (mL); NNaOH is the normality of the NaOH solution; 0.064 is the citric acid conversion factor (g/mmol) and Vsample is the volume of filtrate titrated (mL).

Mold and yeast enumeration

2.10.5

For mold and yeast enumeration, 25 g of cherry tomatoes were aseptically homogenized in 225 mL of sterile 0.1% peptone water using a stomacher (Stomacher 400 circulator, Seward Ltd., West Sussex, UK), for 2 min. Serial decimal dilutions were prepared in sterile 0.1% peptone water. Aliquots (100 μL) of appropriate dilutions were surface-plated on acidified potato dextrose agar (PDA; final pH 3.5, adjusted with 10% tartaric acid) to selectively inhibit bacterial growth. Plates were incubated at 25 °C for 5 days. Mold and yeast colonies were enumerated, and the results were expressed as log cfu/g (Cserhalmi et al., 2006). All microbiological analyses were performed in triplicate.

Firmness

2.10.6

Cherry tomatoes firmness was assessed using a penetrometer (FT 40; Wagner Instruments, Greenwich, CT, USA) on three sides of the fruit surface and the average was taken and the results were expressed as newtons (N) (Zhang et al., 2019).

Statistical analysis

2.11

All data were expressed as mean ± standard deviation from at least triplicate measurements. Significant differences among samples or treatments were evaluated using one-way ANOVA. The significant differences between means were set at p < 0.05.

Results and discussion

3

Effect of enzyme type and E/S ratio on the degree of hydrolysis of CSG

3.1

CSG was successfully extracted and subsequently hydrolyzed using Savinase and Neutrase, two proteases widely used in industrial applications due to their high catalytic efficiency. The DH of the resulting hydrolysates was monitored over 240 min at two enzyme-to-substrate ratios (30 and 50 U/mg protein). As shown in Fig. 1, Savinase showed a higher hydrolytic activity than Neutrase at both E/S ratios. At 30 U/mg, the DH of CSGH_S30 increased rapidly during the first 90 min, reaching 21.52%, and gradually rose to 29.15% after 240 min. In contrast, CSGH_N30 showed a slower and more limited hydrolysis, attaining only 10.37% at 240 min. The raising of E/S ratio to 50 U/mg enhanced hydrolysis for both enzymes, where CSGH_S50 reached a DH of 32.63% at 240 min, and CSGH_N50 attained 15.26%. This pattern confirmed that higher E/S ratios led to more extensive hydrolysis by increasing the availability of enzyme active sites.Fig. 1. Degree of hydrolysis (%) as a function of hydrolysis time. Hydrolysates obtained with Neutrase at 30 and 50 U per mg of substrate are labeled CSGH_N30 and CSGH_N50, respectively, while those produced with Savinase at the same ratios are labeled CSGH_S30 and CSGH_S50. *(U/mg refers to units per mg of substrate, not specific enzyme activity.)*Fig. 1

The observed trends were in line with previous studies showing that both hydrolysis time and enzyme-substrate (E/S) ratio had a major effect on the DH. The rapid increase in DH during the early hydrolysis stage (0–90 min), followed by a slower yet continuous rise, suggested that enzyme activity was initially facilitated by high substrate accessibility before gradually stabilizing as the substrate became limited. Increasing the E/S ratio further amplified DH, which is consistent with the findings of Noman et al. (2018), who reported that greater enzyme availability enhanced active site exposure and accelerated proteolysis. This trend explained the higher DH recorded for Savinase at E/S 50 (32.63%) compared to E/S 30 (29.15%), with Neutrase showing a similar but less pronounced pattern. The non-linear progression of DH observed here was also in agreement with Tabarestani et al. (2024), who reported that hydrolysis time and E/S ratio had interactive and quadratic effects on gelatin hydrolysis kinetics. The consistently superior performance of Savinase over Neutrase could be attributed to differences in enzyme specificity and cleavage mechanisms, as reported by Ketnawa et al. (2017) for catfish gelatin hydrolysates produced by visceral peptidase and bovine trypsin. Overall, these results highlighted that optimized combinations of E/S ratio and hydrolysis duration were essential to maximize hydrolysis efficiency.

Proximate composition of CSGHs

3.2

As shown in Table 1, the proximate composition of CSG and its hydrolysates varied significantly depending on the enzyme used and the E/S ratio. All hydrolysates showed higher moisture contents than native gelatin, with CSGH_N30 presenting the highest value. This increase was generally attributed to the enhanced water-binding capacity of short peptides generated during enzymatic hydrolysis, which were more hydrophilic and therefore better able to retain moisture (He et al., 2025).Table 1. Proximate composition (g/100 g) of CSGHs.Table 1CSGCSGH_N30CSGH_N50CSGH_S30CSGH_S50Moisture6.04 ± 1.129.02 ± 0.02^a^8.15 ± 0.06^b^7.25 ± 0.32^c^6.32 ± 0.32^d^Ash1.55 ± 0.7612.12 ± 0.03^a^11.87 ± 0.01^b^10.12 ± 0.03^c^9. 34 ± 0.03^d^Fat1.57 ± 0.260.49 ± 0.01^a^0.38 ± 0.01^b^0.34 ± 0.02^b^0.25 ± 0.01^c^Protein90.72 ± 1.4577.95 ± 0.12^d^79.16 ± 0.27^c^81.88 ± 0.14^b^83.76 ± 0.2^a^^a,b,c,d^Different letters in the same line indicate significant differences (p < 0.05). Hydrolysates obtained with Neutrase at 30 and 50 U per mg of substrate are labeled CSGH_N30 and CSGH_N50, respectively, while those produced with Savinase at the same ratios are labeled CSGH_S30 and CSGH_S50. (U/mg refers to units per mg of substrate, not specific enzyme activity.)

The ash content also increased markedly after hydrolysis, particularly in Neutrase-treated samples, likely due to the retention or incorporation of mineral salts derived from alkaline agents (e.g., NaOH) added to maintain pH during the reaction, an effect previously reported in gelatin hydrolysis systems. A significant decrease in fat content was recorded in all hydrolysates compared to CSG. Similar reductions were reported for enzymatically produced fish and mammalian gelatin hydrolysates, in which lipid fractions were partially removed during centrifugation and supernatant collection steps (Zeng et al., 2019). This removal not only decreased the residual fat content but also improved the oxidative stability of the final hydrolysates by eliminating susceptible lipids.

Although protein content slightly decreased in hydrolysates, it remained relatively high. This reduction might be linked to fragmentation of proteins into small peptides, which were less efficiently detected by standard protein quantification assays. Among the hydrolysates, those produced with Savinase showed higher protein contents than those obtained with Neutrase, which was consistent with the greater extent of hydrolysis observed for Savinase-treated samples.

Amino acid composition of CSGHs

3.3

Multiple studies showed that enzymatic hydrolysis enhanced functional properties, including antioxidant activities, and that these effects were closely related to AA composition (Cho et al., 2025). The AA profile analysis of CSGHs revealed enzyme-dependent shifts in individual amino acids while maintaining the same total AA content (Table 2). The increases observed in several key bioactive AAs can be mechanistically explained. First, sulfur-containing amino acids like Met (+63% in CSGH-S50) enhanced antioxidant capacity through their hydrogen-donating thioether groups, which scavenged free radicals. Second, basic amino acids including His (+107% in CSGH-S50) and Arg (+19% in CSGH-S50) possessed strong metal-chelating properties, particularly through the imidazole ring of His, which efficiently coordinated Fe^2+^ ions (Wu et al., 2024). Third, branched-chain amino acids (Val/Leu/Ile: +28–47% in CSGH-S50) contributed to metal ion chelation via their amide bonds and carbonyl groups. The pronounced increase in Phe (+43% in CSGH-S50), an aromatic AA with radical-stabilizing π-electrons, further enhanced antioxidant potential (Wang et al., 2025).Table 2. Amino acid composition of CSG and CSGHs (number of AA residue/100 residues).Table 2. Amino acidsCSGCSGH_N30CSGH_N50CSGH_S30CSGH_S50Asx3.23.93.63.33.9Glx1212.912.411.712.6Gly2923.124.122.521His1.452.32.382.643Arg7.288.48.68.6Thr1.92.42.22.41.6Ala88.3388.58.6Pro1413.813.913.113.3Hyp1098.69.49.3Ser1.22.11.721.81.9Tyr0.70.70.60.70.74Val1.81.922.152.3Met1.952.482.622.973.17Lys1.91.891.982.61.79Leu1.72.12.22.292.5Ile1.72.22.22.252.4Phe2.32.93.13.13.3Total (%)100100100100****100The aspartic and glutamic acid contents include, respectively, asparagine and glutamine, where Asx = Asp + Asn and Glx = Glu + Gln. Hydrolysates obtained with Neutrase at 30 and 50 U per mg of substrate are labeled CSGH_N30 and CSGH_N50, respectively, while those produced with Savinase at the same ratios are labeled CSGH_S30 and CSGH_S50. (U/mg refers to units per mg of substrate, not specific enzyme activity.)

These modifications reflected differences in enzyme cleavage specificity. Savinase, known for its broader proteolytic activity, released greater amounts of hydrophobic and aromatic residues than Neutrase, which is in line with the high levels of Met, Phe, and His observed in Savinase-derived hydrolysates (Tabarestani et al., 2024). The effect of the E/S ratio further amplified these changes, aligning with previous reports demonstrating dose-dependent hydrolysis efficiency (Noman et al., 2018).

Although the total AA content remained unchanged, the redistribution toward bioactive AAs agreed with findings in other protein hydrolysates (Cho et al., 2025), confirming that enzymatic hydrolysis preferentially liberated specific residues without altering overall composition. Collectively, these shifts indicated that CSGH-S50 may showed superior bioactivity owing to its enriched profile of hydrogen-donating (Met), (ii) metal-chelating (His, Arg), and (iii) radical-stabilizing (Phe, Tyr) AAs, three categories considered important for antioxidant function (Cho et al., 2025; Sun et al., 2020).

Antimicrobial activity of CSGHs

3.4

The antimicrobial potential of CSGHs was assessed using the disk diffusion method, by measuring the clear inhibition zones (expressed in mm) (Table 3). Overall, CSGH_N50 showed slightly higher antimicrobial activity across most tested strains, particularly against Salmonella enterica (27 mm), Enterococcus faecium (21 mm), and Micrococcus luteus (20 mm), while all hydrolysates showed only moderate activity against the tested fungi (6–9 mm). However, inhibition zones for Savinase-derived hydrolysates were comparable in several cases, indicating that enzyme type and E/S ratio had moderate influence on the activity. A modest increase in antimicrobial effect at the higher enzyme-to-substrate ratio (E/S = 50 U/mg) suggested slightly more efficient release of bioactive peptides. These observations aligned with previous studies; Lima et al. (2019) reported that hydrolysates from weakfish protein showed the strongest antibacterial effects at a low DH (≈ 5%), indicating that specific low-molecular-weight peptides or particular amino acid sequences were especially effective against pathogens such as E. coli and S. aureus. This highlighted the importance of carefully adjusting hydrolysis parameters to enhance peptide antimicrobial performance. Consistent with the literature, the results confirmed that enzymatic hydrolysis released peptides capable of inhibiting microbial growth, with their potency depending on (i) the specificity of the target bacteria, (ii) the characteristics of the native protein, and (iii) controlled hydrolysis conditions, including the enzyme type and reaction parameters (Graikini et al., 2024). In addition, the antibacterial properties of such peptides were mainly influenced by their size and hydrophobic AA content, which facilitated both interaction with and destabilization of bacterial membranes (Sun et al., 2016). Overall activity was also influenced by the balance between hydrophilic and hydrophobic residues, and by the ability of cationic peptides to disrupt bacterial membrane integrity through ion displacement, particularly Mg^2+^ and Ca^2+^ (Sun et al., 2016). Altogether, the observed antimicrobial activity highlighted the broad-spectrum potential of camel gelatin peptides as natural bio-preservatives.Table 3. Antibacterial and antifungal activity of CSGHs expressed as inhibition zone diameters (mm).Table 3CSGH_N30CSGH_N50CSGH_S30CSGH_S50Gram-positive bacteriaBacillus cereus16181214Listeria monocytogenes12141214Micrococcus luteus17201415Staphylococcus aureus14171214Enterococcus faecium18211618

Gram-negative bacteriaEscherichia coli18181416Klebsiella pneumoniae14161315Pseudomonas fluorescens14161010Salmonella enterica23272124Salmonella typhi14151214

FungiBotrytis cinerea6677Fusarium oxysporum8899Hydrolysates obtained with Neutrase at 30 and 50 U per mg of substrate are labeled CSGH_N30 and CSGH_N50, respectively, while those produced with Savinase at the same ratios are labeled CSGH_S30 and CSGH_S50. (U/mg refers to units per mg of substrate, not specific enzyme activity.)

In vitro antioxidant activities of the CSGHs

3.5

A set of four complementary assays (DPPH•, ferric-reducing power, ferrous-ion chelation, and β-carotene-linoleate bleaching) was used to capture different antioxidant mechanisms (radical scavenging, electron donation, metal chelation, and protection in a lipid matrix), providing a mechanistic profile of the hydrolysates.

DPPH• radical-scavenging activity

3.5.1

Fig. 2A shows that all hydrolysates increased scavenging capacity in a concentration-dependent manner and followed the hierarchy CSGH_S50 > CSGH_S30 > CSGH_N50 > CSGH_N30 (CSGH_S50: 91.18% at 5 mg/mL; native CSG ≈ 35.8%). The superior performance of Savinase hydrolysates aligned with previous findings showing that enzymatic hydrolysis enhanced radical-scavenging by releasing low-molecular-weight peptides enriched in aromatic and hydrophobic residues capable of donating protons or stabilizing radical intermediates (Ketnawa et al., 2017; Artwiastia et al., 2023; Li et al., 2025). The levels of Tyr, Phe, and other hydrophobic residues observed in Table 2 supported this hypothesis.Fig. 2. Antioxidant activities of CSGHs measured by (A) DPPH• scavenging, (B) ferric reducing power, (C) Metal chelating, and (D) β-carotene bleaching assays.Fig. 2

Ferric-reducing power

3.5.2

As shown in Fig. 2B, the reducing power closely paralleled the DPPH• trends, with CSGH_S50 showing the highest electron-donating capacity (≈2.03 at 5 mg/mL compared to 0.45 for native CSG). This improvement is consistent with the generation of small peptides containing accessible reducing groups (–OH, –SH) and electron-donating residues such as Tyr and Met, which were more effectively released by Savinase at high E/S ratio (Ketnawa et al., 2017; Artwiastia et al., 2023). Similar relationships between DH, peptide size distribution, and reducing power were reported in shark and catfish gelatin hydrolysates (Ketnawa et al., 2017; Tabarestani et al., 2024).

Ferrous-ion chelating activity

3.5.3

Chelation followed the same sample ranking (Fig. 2C), with CSGH_S50 reaching ∼100% at 2 mg/mL compared to 38.7% for native CSG. This enhancement was in line with known contributions of specific amino acids to metal binding: His (via imidazole coordination), Arg, Gly, and sulfur-bearing residues such as Cys and Met (Sun et al., 2020; Wu et al., 2024). The increased abundance and accessibility of these residues in Savinase hydrolysates (Table 2) indicated that Fe^2+^ chelation was mainly mediated by ligand-metal coordination in small peptides rather than by redox quenching alone.

β-Carotene-linoleate bleaching inhibition

3.5.4

The β-carotene-linoleate assay evaluated antioxidant protection in lipid environments, where peroxyl radicals from linoleic acid accelerated β-carotene bleaching (Dawidowicz & Olszowy, 2015). Fig. 2D shows that CSGH_S50 also provided the highest lipid-phase protection (≈96.6% at 5 mg/mL), greatly exceeding native CSG. This enhanced performance likely derived from peptides combining amphipathicity, favoring interfacial localization in emulsified systems, and radical-stabilizing residues such as Tyr, Phe, and Met, which supported efficient neutralization of lipid-derived peroxyl radicals.

Across all assays the consistent pattern S50 > S30 > N50 > N30 reflected both the extent of hydrolysis and the specificity of Savinase in releasing small, hydrophobic/aromatic, and metal-binding peptides. The results of radical scavenging, reducing power, metal chelation, and lipid-phase protection, strongly supported a multi-mechanistic antioxidant action driven by peptide size, composition, and structure (Ketnawa et al., 2017; Li et al., 2025; Sun et al., 2020; Wu et al., 2024). Overall, the obtained results showed clear links between AA/peptide profiles and functional outcomes, confirming that controlled hydrolysis of CSG produced multifunctional antioxidant hydrolysates with complementary radical-scavenging, ferrous-ion chelating, and lipid-protective capacities, suitable for both aqueous and lipid food systems and potentially contributing to delayed lipid oxidation and microbial growth.

pH stability of CSGH_S50

3.6

CSGH_S50, which showed the highest antioxidant properties, was chosen for further stability studies. The pH stability of CSGH_S50 was essential for its application in food matrices ranging from acidic to mildly alkaline conditions. As shown in Fig. 3A, the antioxidant responses varied with pH, consistent with observations for shark and fish gelatin hydrolysates (Ketnawa et al., 2017; Kittiphattanabawon et al., 2012). The β-carotene-linoleate bleaching inhibition peaked at pH 9 (≈131%), supporting earlier reports that alkaline environments enhanced peptide terminal-group ionization and lipid-peroxidation inhibition (Nasri et al., 2014). The DPPH**•** scavenging remained comparatively stable across pH 1–11 (≈86–118%), consistent with the robustness previously described for catfish gelatin hydrolysates (Ketnawa et al., 2017). The reducing power increased markedly at pH 9 (≈145% of the reference), likely due to pH-induced exposure of electron-donating residues (–SH, –NH₂), a mechanism also reported by Kittiphattanabawon et al. (2012).Fig. 3. Antioxidant stability of CSGH_S50. (A) pH-dependent residual activities, calculated relative to the activity at pH 7 (values >100% indicate higher activity than the reference). (B) Thermal stability: residual activities vs heating time, relative to the untreated control (0 min). (C) Effect of in vitro gastrointestinal digestion: residual activities relative to the undigested control (0 min). Assays in all panels include β-carotene-linoleate bleaching, ferric reducing power, and DPPH• radical-scavenging activity.Fig. 3

This trend aligned with the general principle that pH-driven charge modulation can optimize peptide bioactivity (Korhonen et al., 1998). While activity decreased slightly under extreme alkalinity (pH > 10), consistent with reported destabilization patterns of gelatin hydrolysates (Kittiphattanabawon et al., 2012), CSGH_S50 still retained more than 85% of its scavenging capacity under acidic conditions (pH 1–6). This supported its suitability for low-pH applications, such as fruit beverages, fermented products, or acidic sauces. Overall, these results confirmed that CSGH_S50 maintained broad pH stability, with maximal performance under mild alkalinity (pH 7–9).

Thermal stability of CSGH_S50

3.7

Fig. 3B shows that thermal exposure (100 °C, 240 min) significantly enhanced (p < 0.05) rather than reducedthe antioxidant activity of CSGH_S50, providing strong evidence for its suitability in heat-processed foods. DPPH• scavenging increased by ≈40%, a trend that aligned with the heat-induced enhancement reported for Alcalase-marine gelatin hydrolysates (Jridi et al., 2014). Reducing power showed a particularly strong rise (≈580% at 240 min), comparable to the large increase reported for marine gelatin hydrolysates upon heating (Jridi et al., 2014). Likewise, β-carotene-linoleate bleaching inhibition improved progressively (≈45%), reflecting the retention of peptide efficacy in lipid-oxidation systems even after prolonged heating.

This thermal activation likely resulted from the small size and low hydrophobicity of CSGH_S50 peptides, which limited aggregation, maintained solubility, and facilitated the formation of new antioxidant fragments during heating. Such mechanisms differed from typical protein hydrolysates that undergo thermal denaturation, giving CSGH_S50 an important technological advantage. Together, these results indicated that CSGH_S50 remained thermally stable and even gained activity during extended heating, reinforcing its potential for high-temperature food applications.

Stability in gastro-intestinal model system

3.8

Digestive stability is a key prerequisite for bioactive peptides intended for functional foods, as peptides that remain intact during digestion may be absorbed and exert systemic antioxidant effects. In this study, the stability of CSGH_S50 (10 mg/mL) under simulated gastrointestinal conditions was evaluated to predict its biological behavior. Fig. 3C shows that CSGH_S50 not only remained stable under gastrointestinal digestion but also showed a progressive enhancement of its antioxidant activities, consistent with patterns reported for fish protein hydrolysates (Liu et al., 2010; Jridi et al., 2014; Fu et al., 2015). DPPH• scavenging increased to ≈132% of initial values (p < 0.05), with the largest increase during the gastric phase (0–60 min). This aligned with previous reports showing that pepsin releases antioxidant peptides enriched in aromatic (Tyr, Trp, Phe) and hydrophobic (Leu, Ile, Val) residues (Liu et al., 2010). Reducing power also increased to ≈156% of baseline, especially during the intestinal phase (120–240 min), likely due to enzymatic exposure of electron-donating groups (–SH, –NH_2_) as reported by Fu et al. (2015). Furthermore, β-carotene-linoleate bleaching inhibition showed important improvement (≈120%), supporting the notion that digestion can generate surface-active peptides capable of protecting the lipid-water interface (Jridi et al., 2014).

Collectively, these results supported the emerging consensus that gastrointestinal enzymes often activated rather than reduced the antioxidant potential of protein hydrolysates (Fu et al., 2015). The maintained and even enhanced antioxidant performance of CSGH_S50 throughout digestion indicated promising gastrointestinal stability and potential systemic bioavailability, reinforcing its suitability as a functional food ingredient aimed at mitigating oxidative stress within the digestive tract.

Preservation of cherry tomatoes: Effects on quality attributes using CSG/CSGH_S50 edible coatings

3.9

Weight loss

3.9.1

Fig. 4A shows that weight loss increased progressively for all treatments throughout the 21-day storage period. At day 7, no significant differences (p > 0.05) were detected among the samples, indicating that the coatings had not yet exerted a measurable effect on moisture retention. At day 14, the control group (uncoated) showed a slight but measurable increase in weight loss (≈11%), remaining lower than G1 (glycerol) but higher than both G2 (gltverol + CSG) and G3 (glycerol + CSGH_S50), which confirmed the limited intrinsic barrier of the fruit's natural skin. By day 21, the G1 treatment showed a significantly (p < 0.05) higher weight loss than the other groups, suggesting weaker water-barrier properties when glycerol was used alone, likely due to its hygroscopic nature, which increased water vapor permeability (da Costa de Quadros et al., 2020). At this stage, G1 reached the highest loss (20.8%), whereas the control increased to 17.5%, while G2 and G3 maintained lower values (15.6% and 15.5%, respectively), highlighting the enhanced protective capacity of CSG-based coatings compared to the moderate natural barrier of the uncoated tomatoes.Fig. 4. Changes in cherry tomato quality attributes during storage: (A) weight loss, (B) total color difference (ΔE), (C) pH, (D) total titratable acidity (TTA) and (E) firmness. Treatments were: C, control (uncoated); G1, 3% glycerol; G2, 3% glycerol +2% CSGH_S50; G3, 3% CSG + 2% CSGH_S50.Fig. 4

These trends were visually confirmed by the appearance of the fruits (Fig. 5). Tomatoes coated with G3 maintained a smooth, firm, and homogeneous surface, indicating effective water retention and better structural preservation. However, fruits from the control and particularly G1 developed a wrinkled surface, characteristic of moisture loss and early dehydration. The G2 treatment showed intermediate behavior, with less wrinkling than C and G1, although not as well preserved as G3. Similar observations were reported by Cipolatti et al. (2012), who highlighted the protective role of protein-based edible films against water vapor diffusion. Das et al. (2013) also reported that starch-glycerol-lipid coatings effectively reduced weight loss in tomatoes. Furthermore, Fagundes et al. (2014) and Buendía-Moreno et al. (2019) emphasized that the addition of hydrophobic compounds in coating formulations could further enhance resistance to moisture transfer by forming an additional barrier over the natural wax layer of tomatoes. In contrast, the use of hydrophilic compounds, such as glycerol alone, might have induced the formation of pores or cracks in the coating structure, thereby increased water vapor permeability and accelerated weight loss (da Costa de Quadros et al., 2020).Fig. 5. Cherry tomato appearance during storage. Treatments were: C, control (uncoated); G1, 3% glycerol; G2, 3% glycerol + 2% CSGH_S50; G3, 3% CSG + 2% CSGH_S50.Fig. 5

Overall, these results were consistent with previous reports demonstrating that protein-based edible films, including gelatin, effectively reduced water vapor diffusion and weight loss in fruits (Cipolatti et al., 2012; Das et al., 2013; Fagundes et al., 2014; Ahmed et al., 2023; Buendía-Moreno et al., 2019), whereas hydrophilic compounds like glycerol alone might have increased permeability and accelerated moisture loss (da Costa de Quadros et al., 2020).

Total color difference (ΔE)

3.9.2

Color is a key visual quality attribute of tomatoes and strongly influences consumer acceptance, as it is often the first indicator of freshness, ripeness, and marketability (Kapetanakou et al., 2024). During storage, color deterioration results from moisture loss, oxidative reactions, chlorophyll degradation, and metabolic processes, such as enzymatic browning and pigment oxidation (Fagundes et al., 2014).

Fig. 4B shows that ΔE values increased progressively in all treatments throughout the 21-day storage period, confirming gradual color deterioration. The control group showed the highest color change, reaching ΔE = 13.17 at day 21, reflecting marked loss of visual freshness (Fig. 5). This pronounced shift may have been attributed to higher water loss, increased oxygen exposure, and an accelerated rate of oxidative and metabolic reactions typical of uncoated fruits (Ahmed et al., 2023).

Coated tomatoes showed a markedly slower increase in ΔE. G1 (glycerol) maintained lower values than the control at all storage times (ΔE = 3.84, 6.78, and 10.44 at days 7, 14, and 21), indicating partial protection. However, its hydrophilic nature might have limited its capacity to prevent pigment degradation, consistent with previous reports describing the limited barrier properties of glycerol-rich films (da Costa de Quadros et al., 2020).

The coatings containing CSGH_S50, and especially (G3) were the most effective in preserving tomato color. G2 reached ΔE values of 2.7, 4.64, and 7.21, while G3 maintained the lowest values (2.5, 3.5, and 6.32) throughout storage. These results suggested that protein-based matrices form a more cohesive and oxygen-limiting barrier, reducing transpiration, delaying pigment oxidation, and slowing the enzymatic and non-enzymatic reactions responsible for color loss. Similar findings were reported for gelatin-based coatings applied to fresh produce, where improved color stability was attributed to reduced permeability to oxygen and moisture (Cipolatti et al., 2012; Das et al., 2013; Buendía-Moreno et al., 2019).

Overall, the gelatin-enriched formulation (G3) provided the greatest protection against color degradation, confirming its effectiveness in maintaining visual quality and extending the shelf life of cherry tomatoes. These results highlighted the role of hydrolysate-fortified protein coatings as promising active materials for preserving the appearance of fresh fruits.

Chemical quality (pH and TTA)

3.9.3

The evolution of pH and TTA during storage is a key indicator of tomato freshness, as both parameters reflect metabolic activity, organic-acid degradation, and the early stages of spoilage. As shown in Fig. 4C-D, all samples showed a gradual increase in pH and a concomitant decrease in TTA throughout the 21-day storage period, consistent with the natural ripening process during which organic acids were consumed as respiratory substrates (Kapetanakou et al., 2024).

The control group showed the fastest chemical deterioration, with pH rising from 4.54 to 4.93 and TTA dropping from 0.37 to 0.21%, indicating accelerated metabolism and potential early microbial activity. The literature reported that fungal contamination could contribute to pH elevation by producing alkaline metabolites, thereby hastening spoilage (Chen et al., 2023), which was consistent with the sharper chemical shifts observed in the uncoated fruits.

Among the coatings, G1 also showed a TTA decline and at day 21, reached the highest pH (5.17) and the lowest final TTA (0.20%), demonstrating that glycerol alone did not provide effective biochemical stabilization. In contrast, gelatin coating containing the hydrolysate (G3) offered superior preservation of internal quality. G2 maintained intermediate stability (final pH 4.91; TTA 0.26%), while G3 provided the best protection, ending with the lowest pH (4.80) and the highest acidity (0.28%). These results indicated that gelatin-hydrolysate-based coatings acted as functional barriers to oxygen and moisture, thereby limiting respiration rate, slowing organic-acid consumption, and reducing microbial proliferation, mechanisms widely reported for biopolymer-based edible coatings (da Costa de Quadros et al., 2020).

Overall, the incorporation of CSGH_S50 substantially enhanced the ability of the coating to maintain the chemical integrity of cherry tomatoes, confirming the potential of CSGH-enriched edible films to retard biochemical spoilage and extend postharvest shelf life.

Tomatoes firmness

3.9.4

Fruit firmness is a critical quality parameter for cherry tomatoes, directly influencing consumer acceptance, shelf-life potential, and resistance to mechanical damage during handling and transport. Fig. 4E shows that gelatin coating significantly improved the firmness of cherry tomato fruit during storage. Cherry tomato coated with gelatin enriched by CSGH_S50 had significantly firmer fruit (52,1 N) at the end of shelf-life days followed by G2 and G1 (49 N and 41 N, respectively) compared with uncoated fruit (41.7 N).

In the present study, the hypothesis that the firmness of coated cherry tomatoes is preserved can be supported by the finding that the coating reduced the fruit's respiration rate, thereby slowing its metabolic activity and ripening process. The maintaining of firmness in coated fruits may be attributed to the low weight loss. In accordance with these findings, Buthane et al. (2025) reported that ‘Romanita’ and ‘Tinker’ cherry tomato coated with Aloe vera gel at different concentrations maintained their firmness during storage of 18 days.

Microbiological analysis

3.9.5

The incorporation of natural biopolymers and protein hydrolysates into edible coatings is known to limit microbial proliferation in fresh fruits by reducing surface moisture, oxygen availability, and nutrient diffusion (Fagundes et al., 2014; da Costa de Quadros et al., 2020). Table 4 presents the evolution of mold and yeast counts in coated and uncoated cherry tomatoes during storage. At day 0, all samples showed absence of detectable molds and yeasts, confirming good initial microbiological quality.Table 4. Mold and yeast counts (cfu/g) of cherry tomatoes during shelf life.Table 4. TreatmentsStorage days071421CAbsence5.0 × 10^3^7 × 10^4^1.5 × 10^5^G1Absence2.2 × 10^3^4 × 10^4^1.2 × 10^5^G2Absence<101.0 × 10^3^6.5 × 10^3^G3Absence<10<104.0 × 10^3^Treatments were: C, control (uncoated); G1, 3% glycerol; G2, 3% glycerol +2% CSGH_S50; G3, 3% CSG + 2% CSGH_S50.

During storage, microbial populations increased in all treatments, but the rate of growth differed markedly depending on the coating composition. The control group (C) showed the fastest proliferation, rising from 5.0 × 10^3^ cfu/g at day 7 to 1.5 × 10^5^ cfu/g at day 21, indicating rapid deterioration in the absence of protective barriers. The G1 treatment showed only limited inhibition, with microbial loads increasing from 2.2 × 10^3^ cfu/g at day 7 to 1.2 × 10^5^ cfu/g at day 21, demonstrating that glycerol alone provided minimal antimicrobial protection.

In contrast, coatings containing the hydrolysate alone (G2) showed substantial microbial suppression, with counts remaining very low at days 7 (<10 cfu/g) and 14 (1.0 × 10^3^ cfu/g), followed by a moderate increase to 6.5 × 10^3^ cfu/g at day 21. The G3 treatment, containing CSG enriched with CSGH_S50, showed the highest inhibitory effect, maintaining microbial loads below detectable levels (<10 cfu/g) up to day 14 and reaching only 4.0 × 10^3^ cfu/g at day 21. This represented a reduction of more than one logarithmic unit compared to G2 and nearly two logarithmic units compared to the control.

These findings indicated that gelatin- and hydrolysate-based coatings provided both physical and biochemical protection, limiting moisture accumulation and forming an unfavorable environment for fungal growth. The enhanced activity observed in G3 may have been attributed to the presence of bioactive peptides with antifungal properties, as measured in Table 3, and as previously reported for protein hydrolysates incorporated into edible films (da Costa de Quadros et al., 2020).

Overall, the superior performance of the G3 coating demonstrated the strong inhibitory potential of hydrolysate-enriched edible films, effectively delayed microbial proliferation and extended the microbiological shelf life of cherry tomatoes during storage.

Conclusions

4

This study demonstrated that enzymatic hydrolysis of camel skin gelatin using Savinase and Neutrase successfully generated bioactive peptides, with CSGH_S50 showing strong antioxidant propeties. When incorporated into a gelatin-based coating for cherry tomatoes, CSGH_S50 effectively reduced weight loss, preserved pH and titratable acidity, limited color degradation, and delayed microbial proliferation, thereby extended shelf life. The hydrolysate also showed high stability under a wide range of pH, thermal treatments, and simulated gastrointestinal conditions, suggesting its robustness for practical applications. Overall, these results highlighted the potential of bioactive, hydrolysate-enriched gelatin coatings as a natural and multifunctional strategy to maintain the quality and safety of fresh products during storage.

Future studies should investigate large-scale applications, sensory and organoleptic properties, the interaction with other natural preservatives, and the mechanisms underlying the antimicrobial and antioxidant effects, to further optimize coating performance and facilitate commercial implementation.

CRediT authorship contribution statement

Imen Hamrouni: Writing – original draft, Methodology, Conceptualization. Ola Abdelhedi: Investigation, Data curation. Nasir A. Ibrahim: Writing – review & editing, Validation, Supervision. Walid Elfalleh: Writing – review & editing, Validation, Supervision. Nahed Fakhfakh: Writing – review & editing, Validation, Supervision. Mourad Jridi: Writing – review & editing, Validation, Supervision.

Funding statement

This work was supported and funded by the Deanship of Scientific Research at Imam Mohammad Ibn Saud Islamic University (IMSIU) (grant number IMSIU-DDRSP2601).

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.