The macroalgae utilized in this study is a species of Chondrus crispus from the market. The materials used for protein extraction and fractionation are methanol, ammonium sulfate, sodium dihydrogen phosphate, disodium hydrogen phosphate, sodium citrate, and citric acid purchased from Sigma Aldrich. Enzymes were purchased from G-bioscience, Geno Technology Inc. The solid-phase extraction cartridges were purchased from Thermo Scientific. Protein/peptide concentrations were determined by bicinchoninic acid (BCA) protein assay from Solarbio. The materials used in the antibacterial test are Mueller Hinton Broth and agarose obtained from Himedia, paper disc and streptomycin were obtained from Oxoid. The equipment used in the characterization process is acetonitrile (hyper grade for LC-MS, LiChrosolv), water (MS grade), trifluoroacetic acid (TFA) (Merck), and cellulose dialysis membrane.

Fine dry macroalgae weighing 8 grams was diluted by water 1:25 (g/mL) and sonicated at room temperature and 42 kHz frequency for 1 hour. The mixture was centrifuged for 10 minutes at 2000 rpm. Supernatant was separated from the impurities. Protein was extracted from the supernatant using the salting-in method using ammonium sulphate at the saturated point (80% w/v). The protein pellet was obtained after the mixture was centrifuged for 1 hour at 4500 rpm and 4 °C. Salting out process was carried out overnight using dialyzed cellulose membrane to remove the salts. Protein concentration was determined by BCA protein assay reagent at 562 nm (Walker 1994). The protein pellet was dried by freeze-dry process for 24 hours.

Protein was hydrolyzed using trypsin and chymotrypsin enzymes. The crude protein was dissolved using ammonium bicarbonate to obtain 12 mg/mL solution concentration and heated at 90 °C for 15 minutes. Enzymes were added in ratios 1:10; 1:20; 1:30, 1:40, 1: 50, and 1:60 E:S (w/w), the mixture was vortex for 1 minute and incubated for 20 h at 37 °C at pH 8. Incubation time optimization was determined with 4, 8, 12, 16, 20 and 24-hours variation. The enzyme was inactivated by heating at 90 °C for 15 minutes, supernatant from the mixture was separated by centrifuge at 3000 rpm for 10 minutes. The hydrolyzed protein was filtered using 3 kDa ultrafiltarion. The concentration was determined by BCA protein assay kit at 562 nm. The degree of hydrolysis (DH) was determined using the formula below.

$$DH=\frac{hydrolyzedmassprotein<3kDa}{crudeproteinmass}\times 100\mathrm{\%}$$

Hydrolyzed proteins were fractionated by solid phase extraction method with a strong cation exchanger cartridge (SPE-SCX). The SCX cartridge is prepared by pure water and methanol with a flow rate of 1–3 mL/minute and conditioned by pH 3 buffer. The protein hydrolysate fractionated in citrate and phosphate buffer gradient pH 3; 4; 5; 6; 7; 8 and 9. Peptide fraction purified by SPE with polar enhance polymer (PEP) cartridge and methanol as eluent. Peptide fraction in methanol dried by nitrogen gas and diluted by purified water. Fraction concentration was determined by BCA protein assay kit at 562 nm (Walker 1994).

Antibacterial activity of fraction tested by disc diffusion dilution against Staphylococcus aureus ATCC 25923 from Faculty of Medicine, Universitas Gadjah Mada. Inoculant made by growth bacteria in Mueller-Hinton broth for 24 hours and diluted until the bacteria concentration was 0.1 OD (1 × 10⁸ CFU/mL). The disc diffusion assay was carried out by spreading the inoculant at Muller Hinton agar. Purified water was used as negative control; streptomycin disc 10 µg was used as positive control. The fraction with a concentration of 1000 ppm was added for 10 µL to obtain a 10 µg sample on the disc. The inoculant was incubated for 24 hours at 37 °C. Peptides sequence of the active fraction was identified by liquid chromatography high-resolution mass spectroscopy (LC-HRMS) (Thermo Scientific).

Peptides sequences were identified by liquid chromatography high-resolution mass spectroscopy (LC-HRMS) (Thermo Scientific). Samples were filtered by a 2 µm syringe filter with a 100 Å pore size. Mobile phases were used are 0.05% TFA in water (A) and 0.1% TFA in water/acetonitrile 20:80 (B). The gradient elution program for LC was displayed in Table 1 with 50 µL/minute rate flow. MS/MS analyses were carried out by electron spray ionization (ESI) method with positive ion mode with 250–1.800 m/z range with full-MS/dd-MS2. The resolving power was adjusted at 140.000 FWHM for full-MS, and 17.500 FWHM for dd-MS2. Raw data were analyzed by Proteome Discoverer 2.1 with Sequence HT algorithms using the database Chondrus crispus from UniProt (www.uniprot.org/).

Physical properties analysis and antimicrobial peptide prediction were carried out to selection of active peptide. Scoring of peptides was carried out using the Collection of Anti-Microbial Peptides (CAMPr4) server (www.camp.bicnirrh.res.in). The physical properties of sequences were analyzed by EMBOSS PepStats from the European Bioinformatics Institute (EBI) (https://www.ebi.ac.uk/Tools/seqstats/emboss_pepstats). Toxicity was predicted by ToxinPred (https://webs.iiitd.edu.in/raghava/toxinpred/) (Gupta et al. 2013). The secondary structure of peptides was predicted using Discovery Studio 2021 client software, and the 3D structures were built by de novo PEP-FOLD v4 web server (https://bioserv.rpbs.univ-paris-diderot.fr/services/PEP-FOLD/), with 200 simulation cycle at 370 K (Shen et al. 2014). Helical wheel projection was analyzed by HeliQuest (https://heliquest.ipmc.cnrs.fr/cgi-bin/ComputParamsV2.py) (Gautier et al. 2008).

Molecular docking analysis was performed by High Ambiguity Driven protein-protein Docking (Haddock v.2.4) (Honorato et al. 2021). The 3D structures of potential peptides were built using PEP-FOLD v4 web server (https://bioserv.rpbs.univ-paris-diderot.fr/services/PEP-FOLD/). Protein structure of S. aureus DNA gyrase (6Z1A) and dihydrofolate reductase (DHFR, 1DLS) were downloaded from Research Collaboratory for Structural Bioinformatics (RSCB) Protein Data Bank (https://www.rcsb.org/). Protein structure was cleaned from other compounds using USCF Chimera v1.1.4. The binding affinity of the protein-peptide complex were determined by Protein Binding Energy prediction (PRODIGY) (https://wenmr.science.uu.nl/prodigy/) (Xue et al. 2016).

The extraction of protein from macroalgae was conducted by a conventional method employing sonication and ammonium sulphate saturation to salt out the proteins. The method produced the highest yield compared to high-pressure processing and autoclave pre-treatment. Macroalgae crude protein was obtained with a yield of 13.94 ± 0.94%, less than the result from the previous report with a 35.2 ± 3.9% yield (O’Connor et al. 2020). The protein content obtained during extraction depends on various factors such as the extraction method used, the location of harvest, weather conditions, and the maturity level of the macroalgae(Vieira et al. 2018).

The hydrolysate protein was obtained through the digestion of the protein using trypsin and α-chymotrypsin enzymes. Trypsin and α-chymotrypsin were selected to generate peptides with specific C-terminal and positive characteristics. Trypsin is responsible for cleaving the peptide bonds involving lysin and arginine, while α-chymotrypsin cuts in aromatic residues C-terminal (Gráf et al. 2013; Zhang et al. 2020).

To obtain the maximum degree of hydrolysis (DH), the hydrolysis condition was optimized using two parameters: enzyme-substrate ratio and incubation time. Since antibacterial peptides isolated from natural sources commonly consist of 12–50 residues, the hydrolysates from trypsin and α-chymotrypsin were filtered using a 3 kDa filter to obtain small active peptide (Hancock and Sahl 2006). As shown in Fig. 1, the best enzyme-substrate ratio for both enzymes is 1:10, with a consistent decreasing DH to a higher ratio. Incubation time is also optimized by varying the time, resulting in a consistent increase in DH up to 20 hours, followed by a decrease in 24 hours of incubation. The decrease in DH value could be due to the saturation of the reaction rate caused by inhibition of substrate diffusion or enzyme aggregation (Islam et al. 2021). The result showed that the best incubation time and enzyme-substrate ratio for both enzymes are 20 hours and 1:10. The best DH value exhibited by trypsin-digested hydrolysate was 44.57%. Whereas the best DH value digested by α-chymotrypsin-digested hydrolysate was 54.66%. This result was used to generate new antibacterial peptides from macroalgae Chondrus crispus protein hydrolysate.

Figure 1. 

The degree of hydrolysis of macroalgae Chondrus crispus protein hydrolysate varied by A. Enzyme-substrate ratio (mg/mg) and B. incubation time (hours).

Trypsin-digested and α-chymotrypsin-digested hydrolysate proteins were fractionated by pH variation from 3 to 9 and purified using a PEP cartridge to reduce the salt content from the buffer solution before being tested for their antibacterial activity. Antibacterial activity of peptides fraction was assayed against Gram-negative bacteria S. aureus. Table 2 and Fig. 2 show α-chymotrypsin-digested fractions pH 8 and pH 9 exhibited high inhibition zones of 13.5 and 14.4 mm, respectively, compared to streptomycin as positive control. On the other hand, the remaining fraction from α-chymotrypsin-digested fractions pH 6 and pH 7 showed moderate activity with inhibition zones of 6.5 and 8.0 mm, compared to positive control. Regarding the trypsin-digested fractions, only trypsin-digested fraction pH 4 and pH 6 fractions demonstrated antibacterial activity against S. aureus with small inhibition zones of 2.3 and 5.0 mm, respectively.

Figure 2. 

Inhibition zone of A. trypsin-digested fraction eluted at pH 3; 4; 5; 6; 7; 8; 9; B. α-chymotrypsin-digested fraction eluted at pH 6; 7; 8; 9.

The sequences of antibacterial peptides from all active fractions were identified by LC-HRMS with database of Chondrus crispus protein from the UniProt protein data bank. As shown in Table 3, thirty-one peptides were identified from active fractions. The molecular weight of peptides ranged from 608.70 to 1857.81 Da. Several peptides exhibited high AMPs prediction scores (>0.50), indicating their potential as antimicrobial peptides. All identified peptides

were predicted as non-toxic peptides. Several identified peptides exhibited hydrophobic characteristics, as indicated by their grand average of hydropathicity index (GRAVY) and hydrophobicity values. Positive GRAVY values suggest the peptides are hydrophobic, while negative values signify hydrophilicity (Chang and Yang 2013). Antibacterial peptides with 40–60% hydrophobic residues exhibited good antibacterial activity against broad-spectrum bacteria (Hazam et al. 2023).

The activity of peptides is influenced by residual amino acids, total charge, hydrophobicity, amphipathicity, and secondary structure (Li et al. 2021). Most AMPs are cationic amphipathic peptides with cationic amino acids such as lysine (K), arginine (R), or histidine (H) in large numbers. The mechanism of action of cationic AMPs depends on its ability to interact with the cell walls of microbes (Rehal et al. 2019). Since bacterial cell walls are rich in lipid anions, it is generally believed that positively charged peptides have an interaction with cell membranes (Chen et al. 2017). The interaction leads to bacterial death by cell membrane damage, resulting in cytoplasmic content leakage and cell lysis (Lei et al. 2019). Hydrophobicity is another crucial physicochemical property of antibacterial peptides that aids in penetrating bacteria cell membranes. Hydrophobic residues such as isoleucine (I), leucine (L), tryptophane (W), and valine (V) are critical to secondary structure formation and interaction with hydrophobic component of the lipid (Chen et al. 2007). Several negative charge peptides also were detected in the fraction which good antibacterial activity. Anionic peptides can contribute to activity by using cationic metal salt bridges to interact with bacterial cell membranes (Rana et al. 2018).

According to secondary structure prediction by Discovery Studio 2021 client software, several peptides had α-helical and β-sheet secondary structures (Table 3). The secondary structure of the peptide plays a crucial role that influencing the antibacterial activity of peptides. Besides the amphipathic structure, hydrophobicity and total charge, α-helices, and β-sheet structure are also important for several physicochemical mechanisms to destabilize bacterial cell membranes (Salas-Ambrosio et al. 2021). The secondary structure of peptides is also influenced by environmental conditions. Several AMPs can transform from random coil structure to α-helical conformation in anionic environment like bacterial cell membranes (Cao et al. 2023). The characterization of peptide secondary structure conformation using circular dichroism (CD) spectroscopy is still needed in future studies to display and analyze the influence of the environment on peptide structure.

Among the identified peptides, peptides P01 KKNVTTLAPLVF, P05 ARIASL, P10 RASLAL, P11 TQLTKSF, P16 VFVcVTcSKQW, P22 RGSGLFmR, and P26 MASmmK were classified as cationic antibacterial peptide as their physicochemical properties (Table 3). The peptides had good hydrophobicity with 42.87–66.67% hydrophobic residues. These peptides may influence the activity of fraction as shown in Table 2. Especially for the trypsin-digested fraction at pH 4, only peptide P31 NDYATVSDK as an anionic peptide was identified. Several anionic peptides also reported exhibited antibacterial activity against S. aureus. The peptide MCNDCGA with -1 total charge and 50% hydrophobicity gave good activity against S. aureus with 92.53% inhibition rate (Wang et al. 2023). Although several peptides provided promising physicochemical properties and exhibited good inhibition in fractional form, synthesis peptide antibacterial assay of the identified peptide is needed in future studies to know the activity of single peptide.

As shown in Fig. 3, the helical well and 3D structure of peptide P01 KKNVTTLAPLVF showed the peptide provided two opposing hydrophobic and hydrophilic faces; indicating their amphipathic structure (Jones et al. 1992; Narayana et al. 2020). Peptide Aurein 1.2, a short peptide with 13 residues, was identified to have an amphipathic structure with hydrophilic and hydrophobic faces. The peptide provided good activity against Gram-positive and Gram-negative bacteria (Ramezanzadeh et al. 2021).

Figure 3. 

A. Helical wheel projection of peptide P01 KKNVTTLAPLVF by HeliQuest analysis to show amphipathic structure revealing hydrophobic and hydrophilic faces, color description residues; violet: hydrophilic, yellow: hydrophobic, green: special residue, blue: positive charge residue, pink: amidic residue, grey: aliphatic residue; B. 3D structure of P01 KKNVTTLAPLVF peptide by de novo PEP-FOLD v4 web server.

Basic local alignment search tool analysis (BLASH) using CAMPr4 (http://www.camp.bicnirrh.res.in/ncbiBlast/) was carried out to know the identity and similarity of new peptide with peptide database (Gawde et al. 2022). Among the peptides investigated, only peptide P01 KKNVTTLAPLVF has similar sequence to the previous study with peptide codes CP207 and CP26 as shown in Table 4. with identity and similarity are 64% and 73% respectively. CP207 and CP26 peptides were known as α-helical cationic peptides with +7 total charge, 26 residues, and 46% hydrophobicity with difference in Val13 and Aln13 residues. CP207 peptide displayed a wide-ranging activity against Gram-negative bacteria such as E. coli, P. aeruginosa, and S. typhimurium with minimum inhibitory concentration (MIC) of 2, 3, and 3 µg/ml, respectively (Scott et al. 1999). Different from the CP207 peptide, the CP26 peptide exhibited a broader spectrum of antibacterial activity against Gram-negative and Gram-positive bacteria such as S. aureus and E. coli displaying MIC of 64 and 1 µg/ml, respectively (Scott et al. 1999; Friedrich et al. 2000). Among all peptides, peptide P01 KKNVTTLAPLVF exhibited the most potential physicochemical properties. To study the structure, activity, and mechanism of peptide, synthesis of the peptide chemically may need to be performed in future studies.

The inhibition of bacteria growth by the interaction of peptides with enzyme systems has been proposed as a potential target, although the mechanism of action of cationic antibacterial peptides is known to involve damaging the bacterial cell membrane (Chen et al. 2007; Song et al. 2021). The molecular docking study was still being carried out to demonstrate the potential activity of peptides as antibacterial agents through intracellular mechanisms. In this study, only potential peptides with good amphipathicity, hydrophobicity between 40–60%, no modification, and predicted as AMPs were used for molecular docking study.

Two key enzymes that play crucial roles in sustaining bacteria growth are dihydrofolate reductase (DHFR) and DNA gyrase (Hawser et al. 2006; Khan et al. 2018). Dihydrofolate reductase plays a pivotal role as the principal reduction catalyst in the folic acid pathway. It is responsible for the conversion of dihydrofolate into tetrahydrofolate, utilizing NADPH as a coenzyme factor. This enzymatic process is critical for maintaining folic acid homeostasis and is intimately involved in the de novo synthesis of thymidylate, thereby facilitating cell proliferation. DRFH inhibition profoundly disrupts the biosynthesis of tetrahydrofolate, resulting in an inadequate supply of thymidine, consequently the cessation of DNA biosynthesis, and culminating in apoptotic cell death (He et al. 2020). DNA gyrase is also important in the bacterial growth process. During DNA replication, the enzyme induces negative supercoils into DNA molecules and relaxes the positive supercoils. Inhibition of DNA gyrase involves the stabilization of enzyme and DNA complex intermediate in the catalytic process, causing prevention of the DNA fork movement and transcription (Dighe and Collet 2020).

Molecular docking was performed by Haddock to generate the most accurate protein-ligand complex based on energy stability represented by the Haddock score. According to standard critical assessment of predicted interactions (CAPRI), only complexes with RMSD lower than 2 Å that acceptable for docking study (Lensink and Wodak 2010). The binding affinity of the most accurate complex with similar interaction to the native ligand was determined by the PRODIGY web server to show the best interaction for this study.

In Tabel 5, several peptides exhibited strong interaction with DNA gyrase. Peptides P07 and P21 demonstrated stronger interaction with DNA gyrase than other peptides with binding energy -8.0 kcal/mol, close to native ligand Q52 binding energy (-8.3 kcal/mol). An interaction model between peptide P07 and DNA Gyrase is shown in Fig. 5A; hydrogen bonding interactions were observed in Asp1083; Ser1084, Ser1085; and Glu1088 residues. The rest of the interaction was categorized as van der Waals, carbon-hydrogen bond, and alkyl interaction. Peptide P07 exhibited similar mechanisms as native ligand Q52 of DNA gyrase with hydrogen bond in Asp1083 residue, a key residue for DNA gyrase inhibition (Wang et al. 2023). The native ligand placed the ATP binding domain in the DNA gyrase GyrB subunit. Inhibition of DNA gyrase through GyrB subunit is competitive inhibition in ATP binding domain causing ATP hydrolysis inhibition (Rajendram et al. 2014).

Peptide P20 exhibited the strongest interactions with DHFR with binding energy -7.3 kcal/mol (Table 6). As shown in Fig. 5B, these interactions involved hydrogen bonding with residues Arg28, Asn64, and Tyr22. Moreover, the peptide interacted with residues Asp21 with electrostatic interaction. The rest of the interaction was categorized as van der Waals and carbon-hydrogen bond interaction. As the best-predicted peptide, the P20 peptide could bind to DHFR with a similar mechanism and close binding affinity to methotrexate as a native ligand. Peptide P20 exhibited the same hydrogen bond interaction as methotrexate at Arg28 and Asn64 residues (Fig. 4A). Methotrexate (MTX) is a potent inhibitory of DHFR by disturbing the conversion of dihydrofolate into active tetrahydrofolate (THF). Consequently, the availability of THF and its derivatives becomes limited and decreases the biosynthesis of purines and pyrimidines as DNA and RNA precursors in the cellular proliferation process (Bedoui et al. 2019).

Figure 4. 

Native ligand interaction A. {N}-[(4-chlorophenyl)methyl]-1-[2-(6-methoxy-1,5-napthyridin-4-yl)ethyl]piperidin-4amine (Q52) to DNA gyrase (6Z1A); B. methotrexate to dihydrofolate reductase (1DLS).

Figure 5. 

Interaction model between A. Peptide P20 and DNA Gyrase; B. Peptide P07 peptide and dihydrofolate reductase (DHFR), two key enzymes of Staphylococcus aureus.

The result disclosed that peptides P07 and P21 showed strong interaction with DNA gyrase with hydrogen bonds and other interactions. Peptide P20 also exhibited tight binding to dihydrofolate reductase enzyme by hydrogen bonding and other interactions. The finding indicated that peptides from active fractions may inhibit the growth of S. aureus bacteria by intramolecular mechanisms. However, the inhibitory effect assay of peptides from Chondrus crispus hydrolysate protein on DNA Gyrase and DHFR is needed in future investigation.