Amino acid analysis was conducted using reverse-phase high-performance liquid chromatography (RP-HPLC; Model GmbH Hewlett-Packard, Agilent Technologies Inc., Santa Clara, CA, USA) following the specifications reported by Vázquez-Ortiz et al. (1995). Freeze-dried samples, 6 mg, were homogenized with 100 mL of performic acid. Performic acid was prepared by mixing 1 mL of 30% hydrogen peroxide with 19 mL of 97% formic acid and then maintained in a closed container at room temperature for 2 h. After preparation, the performic acid was cooled to 0°C and added to the sample. Following this, 0.9 mL of cold water was added after keeping the mixture at 0°C for 2.5 h. An aliquot of this mixture (250 µL) was lyophilized for proline (Pro) and hydroxyproline (Hyp) analysis.

Subsequently, the sample was hydrolyzed under pressure at 110°C for 18 h using 6-M hydrochloric acid and sodium thioglycolate (in a 1:1 volume ratio). The hydrolysate was neutralized with 4-N sodium hydroxide. An aliquot part was mixed with an equal volume of 10 mg/mL of L-α-amino n-butyric acid, which served as an internal standard. This mixture was vortexed for 1 min with four parts of potassium borate buffer (pH 10.4) and o-phthalaldehyde (OPA, in a 1:1 volume ratio). Immediately after the preparation, 20 µL of this solution was injected into a reverse-phase column (C18 octadecyl dimethylsilane, 100 × 4.6 mm) coupled to a pre-column (30 × 4.6 mm) packed with the same material. Amino acids were evaluated by employing a fluorescence detector, and the resulting peaks were evaluated using the ChemStation program. The elution process involved two buffers at a flow rate of 1.0 mL/min for 25 min. Fluorescence was measured at wavelengths of 330 nm (excitation) and 418 nm (emission). Peak areas and retention period were compared to those of a mixture of commercial amino acid standards.

An analysis was performed in triplicate, and the results were expressed as the number of amino acid residues per 1,000 residues.

Mesoglea collagen extraction was performed as described in previous studies (James et al., 2023; Nagai et al., 1999) with modifications. All processes were achieved at 4°C. Small pieces of jellyfish mesoglea were suspended in NaOH 0.1-M 1:5 (w/v) solution. The samples were stirred for 24 h and centrifuged at 9,000×g for 30 min. The pellets were rinsed until the pH dropped to 7 and freeze-dried. Afterward, the samples were sequentially treated with acetic acid 0.5 M (1:5 w/v) and pepsin (10 mg/g sample in 0.5 acetic acid; 1:5 w/v). At each step, after stirring for 24 h, the samples were centrifuged (6,000×g for 15 min). The protein solution was then dialyzed against water using a cellulose membrane with a 50-kDa molecular weight cutoff. The collected sample was freeze-dried.

The sample’s collagen concentration was determined by its hydroxyproline by the method adhered to the procedure established by Vázquez-Ortiz et al. (2004) and protein content according to the specification established in the Association of Official Analytical Chemists (AOAC, 2000). The proline and hydroxyproline content were quantified using a Hewlett-Packard HPLC system (model GMBH, Waldbronn, Germany) equipped with a fluorescence detector. The pre-column derivatization method abided the procedure established by Vázquez-Ortiz et al. (2004). In brief, a 0.4-M borate buffer at pH 10.4 was combined with a 250-µL aliquot of lyophilized sample to achieve a total volume of 1 mL. From this solution, 250 µL was mixed with 250 µL of 7-chloro-4-nitrobenzo-2-oxa-1,3-diazole solution (2.0 mg/mL in methanol). The resulting mixture was then heated at 60°C, and precisely 5.0 min later, 50 µL of 0.1-M HCl was added. This mixture was cooled to 0°C for 30 min, after which 10 µL of the sample was manually injected into HPLC system (Varian, Inc., Palo Alto, CA). Amino acid separation was conducted under the same conditions as described previously.

Crude protein was quantified using the Dumas method in accordance with the official AOAC method (993.13), employing a LECO FP-2000 nitrogen/protein analyzer (St. Joseph, MI, USA). This method involves high-temperature combustion to release nitrogen from the sample, which is subsequently measured through thermal conductivity detection. A conversion factor of 5.8 was applied to convert nitrogen (N) content into a percentage (%) of protein in the sample (Doyle et al., 2007).

The molecular mass of collagen extracts from freeze-dried jellyfish mesoglea were analyzed with sodium dodecyl sulphate–polyacrylamide gel electrophoresis (SDS-PAGE) (Laemmli, 1970) using 7% acrylamide separating gel. Protein extract samples and standard collagen type I from calf skin (Sigma-Aldrich) were dissolved in 0.02-M sodium phosphate, pH 7.2, containing 1% SDS and 3.5-M urea (Kimura et al., 1983). Proteins (20 µL, 12 µg) were loaded onto electrophoresis gel, which was developed in Miniprotean^® IIIBio-Rad electrophoretic equipment (Hercules, CA, USA) at 120 V for 90 min. Molecular weight was determined using a protein molecular weight marker kit (Sigma-Aldrich). The gels were analyzed using a densitometer (Bio-Rad GS-700, Hercules).

The thermal analyses of jellyfish mesoglea collagen extracts and calf skin collagen were conducted with differential scanning calorimetry (DSC) using a Perkin Elmer model 8500 DSC system (Perkin Elmer, Waltham, MA, USA). An approximately 6-mg freeze-dried sample was placed in an aluminum pan at a heating rate of 10°C/min from 10°C to 120°C under a nitrogen atmosphere. An empty capsule was used as a reference. The equipment’s software automatically calculated the denaturation temperature (Td) and enthalpy (ΔH) of the jellyfish mesoglea protein extracts. The enthalpy was expressed as joules per g of sample. Analyses were performed in triplicate.

The Fourier transform infrared (FTIR) spectrum of jellyfish mesoglea collagen extracts (0.2 mg in 10-mg potassium bromide) was obtained with a Perkin–Elmer FTIR spectrometer (Frontier MIR/FIR, Waltham, MA, USA), with an average of 16 scans over a spectral range of 4,000–400 cm^-1, at a resolution of 4 cm^-1 (Esparza-Espinoza et al., 2023).

The proton nuclear magnetic resonance (¹H-NMR) spectra of jellyfish mesoglea collagen extracts were obtained using a Bruker Avance 400 nuclear magnetic spectrometer (Billerica, MA) at 400 MHz and 25°C. Freeze-dried samples (1 mg) were dissolved in 0.5 mL of deuterated water (D₂O at 99.6% 2H atom) and 0.5 mL of deuterated potassium hydroxide solution (KOD 40% in D₂O). Dimethylsilapentane-5-sulphonic acid (DSS) was used as a reference, with a 20-ppm spectral window (Esparza-Espinoza et al., 2023).

Three spectra were analyzed for each spectroscopic method to acquire reliable, characteristic and valid results.

Jellyfish mesoglea collagen extracts were dissolved and identified with a nano-LC-MS/MS platform (Ultimate 3000 nano-UHPLC system and Orbitrap Q Exactive HF mass spectrometer with Nanospray Flex ion source; Thermo Scientific, Waltham, MA, USA), as reported previously (Esparza-Espinoza et al., 2023). Briefly, the pepsin protein extracts were poured into a column (C18 SPE, 100Å, 100 × 2 cm, 5 mm; Thermo Scientific) with 0.1% formic acid to remove salt, and 1 mg of the sample was loaded into Nanoflow UPLC with a flow rate of 250 nL/min. The MS/MS scan (300–1,650 m/z, resolution 60,000 at 200 m/z, 3e6) was operated in the top 20 mode using the following settings: resolution 15,000 at 200 m/z, automatic gain control target 1e5, maximum injection time of 19 ms, normalized collision energy at 28%, and an isolation window of 1.4 Th. The charge state exclusion parameters were set to unassigned, 1, >6, and a dynamic exclusion of 30 s. Raw MS files were analyzed and searched against the jellyfish protein database based on the species of the samples using PEAKS Studio 8.5. The parameters were set as follows: the protein modifications were carbamidomethylation (CAM; fixed) and methionine oxidation (variable), and the enzyme specificity was pepsin. There were two maximum missed cleavages; the precursor and ion mass tolerance were 10 ppm and 0.5 Da, respectively.

The antioxidant activity of jellyfish mesoglea collagen extracts and commercial fish marine collagen (0.75 mg/mL) was evaluated by chemical and cellular assays. All measurements were performed in triplicate.

The sample’s 2,2’-azino-bis (3-ethylbenzothiazolin-6-sulphonic acid) (ABTS•+) radical scavenging activity was measured (Re et al., 1999). The working solution was prepared by mixing 7.4-mmol ABTS and 2.45-mmol potassium persulphate. After the ABTS•+ was produced in a dark room at 25°C for 6 h and diluted with methanol, pepsin protein extracts were added. Then, the absorbance was taken at 734 nm using a 96-well microplate reader (Thermo Fisher Scientific Inc., Multiskan GO, New York, NY, USA).

The ferric reducing antioxidant power (FRAP) assay was performed according to Benzie and Strain’s (1996) method, adapted to microplate equipment. The working solutions consisted of a 300-mM acetic acid–sodium acetate buffer (pH 3.6), 20-mM FeCl₃•6H₂O, and 10-mM 2,4,6-tri(2-pyridyl)-s-triazine (TPTZ) in 40-mM HCl (10:1:1 ratio). An aliquot of 20 mL of the samples was mixed with 280 mL of FRAP working solution and incubated at 25°C in the dark for 30 min before its absorbance was recorded at 638 nm using a microplate reader.

The ability of jellyfish collagen extracts to quench the oxygen radicals generated by 2,2’-azobos(2-amidinopropane) (AAPH) was measured with the oxygen radical absorbance capacity (ORAC) technique (Prior et al., 2003). The inhibition of fluorescence decay after mixing collagen extracts (100 mL) with 75-mM phosphate buffer, pH 7.3 (1.7 mL), 0.36-M AAPH (0.1 mL), and 0.048-mM fluorescein (100 mL) was evaluated for 60 min at 37°C at 485 nm (excitation) and 520 nm (emission) using a Cary Eclipse spectrophotometer (Agilent Technologies, Mexico City, Mexico).

The results for ABTS, FRAP, and ORAC were expressed as micromoles of Trolox equivalent per gram of the sample (mmol TE/g). A standard curve was prepared using different Trolox concentrations.

Jellyfish pepsin collagen extracts’ anti-hemolytic capacity against AAPH was evaluated as described by Lu et al. (2010). Briefly, a pure solution of erythrocytes was resuspended in 10-mM phosphate-buffered saline (PBS), pH 7.4 (ratio 5:95, v/v). Three different mixtures were prepared: (1) resuspended erythrocytes (100 mL) + protein extracts (100 mL) + AAPH (100 mL); (2) resuspended erythrocytes (100 mL) + AAPH (100 mL) + 100-mL PBS; and (3) resuspended erythrocytes (100 mL) + 200-mL PBS. After the mixtures were incubated in a dark room at 37°C for 3 h, PBS (1 mL) was added to the solution and centrifuged (1,500×g, 10 min, 4°C). In all, 300 µL of supernatant was transferred to 96-well microplates, and absorbance was read at 540 nm using a spectrophotometric microplate reader. The intact erythrocytes were established, and the results were expressed as percentage inhibition of hemolysis.

The electrophoretic pattern of S. meleagris jellyfish pepsin collagen extracts, under the electrophoretic conditions employed, revealed two “a” chain-like components with a very similar migration and a molecular mass of about 117 kDa (Figure 1, lane B). The migration of jellyfish pepsin collagen extracts was like that of calf skin collagen (Figure 1, lane C), specifically type I a1, but not a2, which may be due to variations in amino acid composition. Similar findings were reported for Chrysaora spp. (Barzideh et al., 2014). This behavior could be attributed to the possibility that both chains originated from distinct molecular collagen types (Kimura et al., 1983). It is observed that some jellyfish collagens are comparable to collagen type I (Rastian et al., 2018). In contrast, others align more closely with type II (Barzideh et al., 2014), and some are related to types IV and V (Lueyot et al., 2021).

Furthermore, a high molecular weight band was detected at around 199 kDa in S. meleagris and above 200 kDa in calf skin, which was associated with the b collagen component. It is established that “α” chains build trimeric molecules, which coil together into a triple-helical conformation. In contrast, “β” chains reveal the presence of an oligomer with a higher degree of order (Eyre et al., 2019). Then, the differences in “α” and “β” chains imply that the analyzed collagen possessed a different degree of cross-linking. Additionally, the presence of faint bands may arise from residual non-collagen proteins, a finding further validated by proteomic analysis, as discussed below.

The amino acid profile of jellyfish pepsin collagen extracts was assessed using HPLC in conjunction with amino acid standards (Figures 2A and 2C). The use of OPA for the separation of amino acids facilitated the identification and quantification of 15 different amino acids (Figure 2B). The chromatograms for proline and hydroxyproline within the jellyfish collagen extract are presented in Figure 2D.

The individual content of free amino acids in jellyfish pepsin collagen extracts and the data expressed as residues/1,000 residues, presented in Table 1, are the average values from a triplicate analysis. Glycine (Gly) was the major amino acid with 335 residues per 1,000. Other high-content amino acids were glutamic acid (Glu, 99 residues per 1,000), alanine (Ala, 83 residues per 1,000), proline (Pro, 8 residues per 1,000), and aspartic acid (Asp, 76 residues per 1,000). Hydroxyproline was also detected in considerable amount (49 residues per 1,000). These data indicate that collagen is the major protein in pepsin collagen extracts. Like Chrysaora spp. (Barzideh et al., 2014) and Rhopilema esculentum (Sudirman et al., 2023), histidine was not found (Table 1).

From the hydroxyproline content of collagen extracts, it was estimated that the collagen content was approximately 45.6%. Similar values were reported for the same jellyfish species (46.4%) (Nagai et al., 1999) as well as Rhizostoma pulmo (47%) (James et al., 2023), but Chrysaora spp. was reported to have 2.4 times less collagen (19%) (Barzideh et al., 2014).

The amino acid composition of jellyfish pepsin collagen extracts was compared to that of calf skin collagen type I as well as the values reported by other authors (Table 1). S. meleagris collagen extracts were determined to have lower Pro and Hyp (p < 0.05) than calf skin type I collagen. Moreover, S. meleagris collagen extracts showed 1.1 times higher Gly than that of S. meleagris exumbrella (Nagai et al., 1999), Chrysaora spp. umbrella (Barzideh et al., 2014), and R. esculentum (Sudirman et al., 2023), but 1.2 times low values of imino acids than Chrysaora spp. umbrella (Barzideh et al., 2014) and R. esculentum (Sudirman et al., 2023). Low contents of Pro and Hyp imply that the secondary structure of collagen is represented more by a b-turn/b-shift structure than triple helices (Derkach et al., 2019). The large number of a b-structure can affect the functional properties of the molecule (Karim and Bhat, 2009) as well as its thermal behaviors (Barzideh et al., 2014), as discussed below.

The differences detected among jellyfish’s amino acid profile and collagen content may be attributed to specific living conditions (Cadar et al., 2023) and the fishing region’s marine environment (Haard et al., 1982).

Differential scanning calorimetry determined the thermal stability of jellyfish pepsin collagen extracts and calf skin collagen. The denaturation temperature (Td) obtained from the heating scan of jellyfish pepsin collagen extracts was 25.1±1.4°C, which was 1.7 times lower (p < 0.05) than that of skin calf type I collagen (41.8±1.3°C). Comparing the data with those reported by other authors, Td values were shown to be similar to collagen extracted from S. nomurai mesoglea (27°C) (Kimura et al., 1983) and S. meleagris exumbrella (26°C) (Nagai et al., 1999) but 1.5 times lower than that of Chrysaora spp. collagen (37.38°C) (Barzideh et al., 2014) and 1.7 times for Aurelia aurita collagen (43.7°C) (Balikci et al., 2024). In addition, jellyfish pepsin collagen extracts exhibited a degree of hydrolysis (DH) of 0.078 J/g, which was 67 times lower (p < 0.05) than that of calf skin collagen (5.26 J/g). Moreover, the DH values for S. meleagris were 23 times lower than Aurelia aurita collagen’s (1.806 J/g) (Balikci et al., 2024) and 30 times lower than Chrysaora spp. collagen (2.35 J/g) (Barzideh et al., 2014). As mentioned, S. meleagris jellyfish pepsin collagen extracts had lower imino acid content than the collagen from calf skin (p < 0.05) and Chrysaora spp. umbrella (Barzideh et al., 2014). The lower DH value of S. meleagris mesoglea pepsin collagen extracts could be related to the low stability of the molecule, which could be due to the hydroxyproline and proline content of its structure (Rodríguez et al., 2017) as well as the method of extraction, water content, and the source organism’s environmental habitat (Safandowska and Pietrucha, 2013).

The jellyfish pepsin collagen extracts’ FTIR spectrum (Figure 3) was comparable to that of other jellyfish species (Table 2). Jellyfish pepsin collagen extracts’ FTIR spectrum displayed five major peaks of collagen characteristics. The peak associated with the N–H stretching frequency (amide A) was observed around 3,290 cm^-1. The N–H stretching vibration usually occurs at 3,440–3,400 cm^-1, and it is reported that when the band position appears at a lower frequency, it suggests that N–H groups are involved in hydrogen bond formation, which can help stabilize the collagen triplehelix (Doyle et al., 1975). The band observed at 2,920 cm^-1 was related to the asymmetric stretch of CH₂ and NH₃⁺(amide B) (Zhang et al., 2014). Amide I, II, and III waves associated with collagen structure (Riaz et al., 2018) were observed from 1,660 cm^-1 to 1,200 cm^-1. CO stretching (amide I) was detected at 1,660 cm^-1 whereas N–H and C–N torsional vibrations (amide II) were observed at 1,585 cm^-1. The absorption band around 1,220 cm^-1 (amide III) is linked to residual CH groups. The triple bands observed around 1,350–1,200 cm^-1 were caused by specific collagen fingerprint tripeptides (Gly-Pro-Hyp) (Riaz et al., 2018).

The spectra of collagen type I from calf skin showed that peaks of amide I and II had a higher frequency (Table 2). The shift of these peaks to lower frequency observed in jellyfish collagen extracts implied that the molecules had a lower degree of molecular order (Payne and Veis, 1988). Moreover, the index absorption ratio of amide III band at 1,450 cm^-1 (amide III/A 1,450), associated with a helical structure and detected in the jellyfish collagen extracts, was lower (0.97) than the ratio from skin calf collagen (1.13). These values implied that fewer intermolecular cross-links were observed in jellyfish collagen extracts than that in calf skin collagen (Chen et al., 2016). Fourier transfom infrarred (FTIR) correlated positively with Td and DH data.

A comparison of S. meleagris collagen’s FTIR amide bands to those reported previously for other jellyfish collagen (Table 2) indicated that most bands obtained in our study were similar to those reported from different jellyfish species, with only one exception, amide A, which was detected at a lower wave number. The lower wave number of the amide A band, which is very sensitive to hydrogen bond strength, implies fewer hydrogen bonds from -NH groups (Balikci et al., 2024). This value indicates that the -NH groups of S. meleagris mesoglea collagen are involved with other groups through hydrogen bonds to stabilize the collagen’s helical structure (Duan et al., 2009; Wang et al., 2007).

The ¹H-NMR spectroscopy is used to detect collagen’s triple helix structure (Baum and Brodsky, 1997). The ¹H-NMR spectrum of S. meleagris’s collagen showed the characteristic resonance of a collagen-like triple helix (Figure 4). The chemical shift at 3.58 ppm indicated the presence of GlyCαH (Acevedo-Jake et al., 2015) whereas the chemical displacements at 3.55 ppm and 3.10 implied the presence of proline protons bound (ProCdH1,h) (Ovando-Roblero et al., 2023). Meanwhile, the resonance at 2.08 ppm was attributed to hydroxyproline protons (HypCbH1) (Ovando-Roblero et al., 2023).

The intense band at 4–6 ppm indicated the presence of water. Collagen fibers contain ≤0.5-g water/g collagen; this water represents the “bound” water fraction, which interacts with collagen’s surface (Fullerton et al., 2006). Chemical displacements at 7.6–7.1 ppm indicated the presence of aromatic protons of pyrrolidine (Sell and Monnier, 1989).

The above observations were consistent with the profiles of amino acid content discussed earlier. The analysis of NMR spectrum, combined with insights from FTIR spectrum, provided strong evidence for the presence of functional groups associated with collagen in extracts. It confirmed the presence of components related to collagen, thereby enhancing our understanding of the structural properties and potential biological significance of the extracted material.

The proteins determined by nano-LC-MS/MS from S. meleagris collagen extracts, summarized in Table 3, are from 3,086 identified spectra (peptide spectrum matches [PSM]), with 707 different peptides belonging to 10 proteins, one of which is collagen type IV. The bioinformatic analysis showed that the collagen detected in this study contained 15 unique peptides. Through LC-MS/MS, the proteomic analysis of gelatin from dried minced jellyfish (Lobonema smithii) showed 12 proteins, in which several types of collagen were detected, including collagen type IV alpha 4 chain, collagen alpha 2 (IX chain), collagen type V alpha 2 chain, collagen 1 alpha 2 chain, and collagen alpha-2 (IX chain) (Lueyot et al., 2021).

Jellyfish collagen has a great degree of chemical simplicity, leading to multifaceted and adaptable tissues. Consequently, it has been categorized as collagen type 0 (Faruqui et al., 2023). It is similar in chemical composition to various collagen types (Faruqui et al., 2023) and shares many functions with more specialized collagens (Bowen et al., 2022), such as bioactive (antioxidant or antimicrobial) or functional properties (such as those exhibited by emulsifiers, foaming substances, colloid stabilizers, and biodegradable film-forming materials) (Chiarelli et al., 2023).