Introduction

Epidemics and pandemics in recent years have proven that public health is strongly under global threat and therefore new and effective drugs in combating these emerging infectious diseases is an urgent requirement. World Health Organization (WHO) reports antibiotic resistance as a serious threat mounting worldwide and hence cautions a period in near future in which infections can no longer be mitigated with the antibiotics currently in use (Xie et al. 2017). The same scenario is being faced not only in human health sector, but also in animal husbandry and aquaculture (Schar et al. 2021). Drug-resistant Vibrio parahaemolyticus strains have been reported in China and South Korea (Loo et al. 2020). Over the last two decades, naturally evolved molecules, the Antimicrobial peptides (AMPs) have been recognized as promising antibiotic-alternatives capable of overcoming antibiotic resistance (Erdem Büyükkiraz and Kesmen 2022). About 40 antimicrobial peptides are presently undergoing clinical trials among which nisin, gramicidin, polymyxins, daptomycin and melittin are in clinical use (Dijksteel et al. 2021).

Marine invertebrates present a wide repertoire of different mechanisms and biomolecules, which enables them to survive in their natural environment as they are directly exposed to microorganisms. As these organisms lack immunological memory, they mainly depend on the innate immune system, comprising humoral and cellular responses (Tincu and Taylor 2004). Among marine invertebrates, crustaceans are the largest and most diverse animal group, abundant and spread in all aquatic habitats, from hypersaline to freshwater ecosystems. As they are constantly bathed in an environment with frequent exposure to both commensals as well as opportunistic pathogens, crustaceans defend themselves by a combination of nonadaptive antimicrobial and antiviral responses guided by haemocytes, the circulating immunocompetent cells (Kulkarni et al. 2021) as well as epithelial cells lining different organs, such as gills and intestine (Silveira et al. 2018; Alenton et al. 2019). It is well known that crustacean AMPs are constitutively synthesised and stored in cytoplasmic granules of specific haemocyte populations (Destoumieux et al. 2000; Vu et al. 2018). Since the discovery of the first crustacean AMPs (Schnapp et al. 1996; Destoumieux et al. 1997), several genetically-encoded AMP families and other antimicrobial-related molecules have been known and characterized at molecular level, especially in decapods due to their commercial importance. Till date, 12 gene-encoded AMP families are recognised in crustaceans i.e., crustins, anti-lipopolysaccharide factors (including crab scygonadins), penaeidins, stylicins, proline-rich AMPs (including the 6.5 kDa bac-like peptide), glycine-rich AMPs, arasins (including the callinectin peptide), hyastatins, astacidins, panusins (including the homarin peptide), paralithocins and armadillidins (Matos and Rosa 2022).

Chisholm and Smith (1992) identified certain antibacterial factors in granular haemocytes of shore crab, Carcinus maenas and Relf et al. (1999) isolated the bioactive peptide, crustin from the granular haemocytes of C. maenas, showing activity against Gram-positive marine bacteria. This was named ‘carcinin’ (Smith and Chisholm 2001) which became the prototype for the later identified group of similar peptides, ‘crustins’, from other crustaceans i.e., Litopenaeus vannamei and L. setiferus (Bartlett et al. 2002). Since then, several crustin like sequences have been reported from various crustaceans such as Penaeus subtilis (Rosa et al. 2007), Hyas araneus and Paralithodes camtschaticus (Sperstad et al. 2009), Fenneropenaeus indicus (Antony et al. 2010; Sruthy et al. 2017), P. monodon (Antony et al. 2011), Scylla serrata (Afsal et al. 2011), Macrobrachium rosenbergii (Huang et al. 2016) and Rimicaris sp. (Wang et al. 2021a, b). Crustins are 7–22 kDa antibacterial proteins in the crustacean hemolymph with the WAP domain containing protein superfamily (Smith et al. 2008; Wang et al. 2021a, b). They have a distinctive signature of 12 cysteine residues out of which four form the cysteine-rich region, while the remaining eight participate in four disulphide bonds, recognized as the 4-disulphide core domain/whey acidic protein domain (4DSC/WAP). The central cysteine of the 4DSC domain (C7–C11 of the crustin signature) seems to be important to maintain the structure (Vargas-Albores and Martínez-Porchas 2017). This three-dimensional arrangement is also found in other proteins displaying different biological functions i.e., antimicrobial activity, proteinase inhibition (Ota et al. 2002) and tissue differentiation (Ranganathan et al. 1999). Contradictory to the conservative C-terminal WAP domain, the N-terminal region of crustins is variable, based on which crustins were classified i.e., Types I–III (Smith et al. 2008). Currently, this classification system also includes proteins containing two WAP domains exhibiting antimicrobial activity (Type IV or double WAP domain-containing proteins) and crustins from hymenopteran insects (Type V) (Matos and Rosa 2022). The present study focuses on molecular and functional characterization of a novel crustin isoform (MC-Crustin) from the gill of the mud crab, Scylla serrata using in silico approach. Scylla serrata mainly inhabits estuaries, mangrove swamps and sheltered coastal habitats and are usually found in muddy bottom. This species is considered as one of the most popular seafood item in South-East Asian countries due to its delicious nature and high demand in international markets (Chandra 2012; Pripanapong and Tongdee 1998). Mud crab farming/fattening has been very attractive and thereby the health management becomes important, necessitating an understanding of their defence potential. In this scenario, a preliminary prediction of properties and potential of MC-crustin was carried out to unravel one of the most important component of its innate immunity, the antimicrobial peptides that confer protection to the animals.

Materials and methods

Sample collection

Mud-crab Scylla serrata (Fig. 1) was collected from a local fish market and transported to lab live. Gills of the crab were excised and macerated thoroughly in TRI Reagent® (Sigma-Aldrich) and preserved at − 20 °C in a Freezer (Sanyo, Japan) prior to RNA isolation.

Fig. 1
figure 1

Green mud-crab, Scylla serrata used in the study

Full size image

Total RNA isolation and cDNA synthesis

Total RNA was isolated using TRI Reagent® (Sigma-Aldrich) as per manufacturer’s protocol and dissolved in DEPC treated water. After confirming purity and quality of RNA in a 0.8% agarose gel, single stranded cDNA was synthesized. For this, only RNA that showed A260:A280 ≥ 1.8 was used. First strand cDNA was generated using 5 μg total RNA, 1 × RT buffer, 2 mM dNTP, 2 mM oligo d(T20), 20U of RNase inhibitor, and 100U of MMLV reverse transcriptase. cDNA synthesis was performed at 42 °C for 1 h followed by an inactivation step at 85 °C for 15 min. Quality of cDNA was tested by PCR amplification of internal control β-actin, primed by crustacean specific β-actin primers (F—5′ CTTGTGGTTGACAATGGCTCCG 3′ and R—5′ TGGTGAAGGAGTAGCCACGCTC 3′).

PCR amplification, TA cloning and sequencing

Crustin was amplified from the S. serrata gill cDNA using crustinF and crustinR primers (crustinF—5′ GAGAGCAGAATTAGACACTGT 3′, crustinR—5′ ATATAGTATAACATAACCATACTTC 3′) designed by Afsal et al. (2013) based on consensus sequence regions of crustins from GenBank. The reaction mixture which included EmeraldAmp® PCR Master Mix, template, primers and water was subjected to 95 °C for 2 min followed by 35 cycles of 94 °C for 15 s, 57 °C for 30 s, and 72 °C for 30 s and a final extension at 72 °C for 10 min. The PCR amplicons were then ligated to pGEM®T Easy vector (Promega) and transformed to DH5α E. coli competent cells, using manufacturer’s protocol. Transformed clones were cultured in LB broth (500 μl), at 37 °C with continuous shaking at 150 rpm for 1 h 30 min and subsequently plated onto LB agar plates containing Ampicillin (100 μg ml−1), X-gal (80 μg ml−1) and IPTG (100 mM). Positive colonies were identified by blue/white screening and the colonies that remained white due to insertional inactivation of the alpha-subunit of beta-galactosidase were selected and cultured overnight in LB broth containing antibiotic (Ampicillin 100 μg ml−1) with continuous shaking at 200 rpm and 37 °C. Plasmids were isolated with GenElute™ Plasmid Miniprep Kit (Sigma), followed by quality confirmation on a 0.8% agarose gel. The recombinant plasmids were subjected to PCR amplification using vector-specific and gene-specific primers. The plasmid was sequenced using T7F and SP6R primers on an ABI 3730XL DNA sequencer (Applied Biosystem, USA) at AgriGenome Sequencing Facility, India.

Sequence analysis and molecular property prediction

The sequence was subjected to BLAST search at NCBI (https://blast.ncbi.nlm.nih.gov/) and the peptide was identified based on similarity score and the sequence was submitted to NCBI BankIt (https://www.ncbi.nlm.nih.gov/WebSub/). ExPASy translate tool (https://web.expasy.org/translate/) was used to deduce amino acid sequence from the nucleotide mRNA sequence. Signal peptide and propeptide sites were denoted by SignalP 5.0 (http://www.cbs. dtu.dk/services/SignalP/) and ProP 1.0 (http://www.cbs. dtu.dk/services/ProP/) (Armenteros et al. 2019; Duckert et al. 2004). Presence of specific peptide domain was predicted using SMART (http://smart.emblheidelberg.de/) (Letunic and Bork 2018). ScanProsite (https://prosite.expasy.org/scanprosite/) provided the cysteine disulphide bonds in the identified domain (Castro et al. 2006). APD3 (http://aps.unmc.edu/AP/main.php) and ProtPram tool (https://web.expasy.org/protparam/) of ExPASy were used to compute various physico-chemical characteristics of the peptide (Wang et al. 2016; Gasteiger et al. 2005). Dipeptide compositions of the peptide were analysed by Pfeature (https://webs.iiitd.edu.in/raghava/pfeature/) (Pande et al. 2019). Hydropathy plot was computed using Kyte and Doolittle by ExPASy ProtScale (https://web.expasy.org/protscale/). Jalview Version: 2.11.1.0 and Mega X applications were used for clustal alignment and phylogenetic analysis respectively (Waterhouse et al. 2009; Kumar et al. 2018). Secondary structure was analyzed in RaptorX (Wang et al. 2010). 3D structure for the peptide was predicted from SWISS-MODEL workspace, Expasy (https://www.expasy.org/resources/swiss-model) and the PDB data generated was visualized in DeepView/Swiss-PdbViewer 4.1.0 (http://www.expasy.org/spdbv/) to compute disulphide bonds, distribution of cationic and hydrophobic residues as well as to predict the electrostatic potential distribution in the 3D model of the peptide (Waterhouse et al. 2018; Guex and Peitsch 1997). PDBsum (http://www.ebi.ac.uk/thornton-srv/databases/pdbsum/Generate.html) provided further structural analyses and Ramachandran plot for the predicted model (Laskowski and Thornton 2022). TANGO (http://tango.crg.es/) and AGGRESCAN (http://bioinf.uab.es/aggrescan/) were used for in vitro and in vivo aggregation analyses (Rousseau et al. 2006; Fernandez-Escamilla et al. 2004; Linding et al. 2004; Conchillo-Solé et al 2007). FoldAmyloid (http://bioinfo.protres.ru/fold-amyloid/) predicted amyloidogenic regions in the peptide (Garbuzynskiy et al. 2010).

Bioactive potential prediction

Online platforms used for prediction of various bioactive potentials of MC-crustin is given in Table 1.

Table1 Software Tools used for bioactive potential prediction (in silico) of MC-Crustin
Full size table

Results

A 500 base pair (bp) amplicon was obtained using crustinF and crustinR primers from the cDNA of gill sample of the green mud crab, Scylla serrata (Fig. 2.).

Fig.2.
figure 2

500 bp amplicon using crustinF and crustinR primers. Lane 1–100 bp ladder, Lane 2–500 bp amplification

Full size image

The sequence analysis of the cloned amplicon revealed it to contain an open reading frame of 336 base pair (bp), encoding 111 amino acid (aa) residue crustin peptide (NCBI Accession number ON513450). With 100% query coverage, the peptide displayed varying range of identities (98.2–73.87%) with previously reported crustins from crabs viz., Scylla olivacea (QJW82625.1) 98.20%, Scylla tranquebarica (AFI56572.1) 98.20%, Scylla serrata (ADW11096.1) 94.59%, Portunus pelagicus (AFN37210.1) 89.19%, Scylla paramamosain (ABY20727.1, ABY20728.1) 90.9% and Portunus trituberculatus (ACM89167.2) 73.87%. A signal peptide cleavage site processed by signal peptidase 1 (Sec/SPI) was predicted between positions Ala21 and Ser22 with N-region Met1–Lys4, H-region Ile5–Val16 and C-region Ala17–Ala21. There were no propeptide cleavage sites identified in the peptide. A transmembrane helix region was predicted from region Ile5–Tyr27. SMART analysis annotated a 4DSC (4-disulphide core)/WAP (Whey Acidic Protein SM000217) domain from His60–Glu109 (Fig. 3).

Fig. 3
figure 3

cDNA sequence (336 bp) (GenBank Accession number-ON513450) encoding crustin antimicrobial peptide (111aa) from the green mud crab Scylla serrata. The translated peptide sequence is shown in colour. The Italicised amino acids form the signal peptide comprising the N-region (yellow), H-region (blue) and C-region (green) with signal peptidase cleavage site predicted at A21–S22 (arrow mark). The remaining sequence comprise of the mature peptide (grey) of MC-crustin within which the SMART annotated WAP/4DSC domain is shown in bold. ‘*’ denotes stop codon. Region forecasted as transmembrane helix is underlined

Full size image

Physico-chemical properties and domain annotation

Molecular weight of the 90 aa, cysteine rich (13%), mature peptide region (MC-crustin) was 10.164 kDa with a theoretical pI of 8.27. Net charge of the peptide accounted to be + 4.25 with 12 (Arginine, R + Lysine, K) positively charged residues and 9 (Aspartate, D) + Glutamate, E) negatively charged residues. Amino acid composition is given in Fig. 4, showing the peptide to be cysteine (C) rich (13%) followed by proline (P) (11%). GRAVY (Grand average hydropathy value) was predicted as 0.45, Wimley–White whole-residue hydrophobicity of the peptide, 17.18 kcal/mol. and the APD defined total hydrophobic ratio as 36%. The ExPasy hydropathy plot also denoted major portion of the peptide falling in the hydrophilic region (Fig. 5). Boman index of the mature peptide was 1.79 kcal/mol. The estimated half-life of peptide was 1.9 h (mammalian reticulocytes, in vitro), > 20 h (yeast, in vivo) and > 10 h (Escherichia coli, in vivo). The instability index (II) of MC-crustin mature peptide was 56.10% which was higher than the instability index of the full-length peptide including the signal peptide, which had the II value of 48%.

Fig. 4
figure 4

Amino acid composition of MC-Crustin from Scylla serrata (calculated by APD3 server http://aps.unmc.edu/AP/main.php)

Full size image
Fig. 5
figure 5

ExPasy plotted hydropathy plot for mature peptide MC-crustin using the Kyte and Doolittle

Full size image

MC-crustin is a Type I crustin, observed with four conserved cysteine residues in the region between the signal peptide and SMART annotated C-terminus WAP domain. MC-crustin has a higher percentage of proline (P) residues (16%) within this region. The WAP motif alone comprises of 50 amino acids with eight cysteine residues, following the conserved arrangement across different types of crustins as well as other WAP domain containing proteins, displaying four disulphide bridges, referred to as the four-disulphide core domain (4DSC). ScanProsite predicted disulphide bonding pattern in WAP domain as C1–C6, C2–C7, C3–C5 and C4–C8. The arrangement of eight cysteines of the WAP domain and the other four cysteine residues outside to it displayed the ‘crustin signature domain’ for MC-crustin as in other crustins (except SWD Type III crustins) which have only WAP domain.

Sequence alignment and phylogenetic analysis

Alignment of crustins revealed the signal sequence of MC-crustin to be Valine (V) rich. However, the signal peptide sequences vary across species and types of crustins, with lengths varying from 16 to 24 aa. The region between the WAP domain and signal peptide marks the types of crustins. Type I crustins including MC-crustin in the present study has a cysteine (11%) rich region with four conserved cysteine residues. Type II crustins comprise of an additional glycine (G) rich region prior to the cysteine rich region. Type III crustins do not contain any sequences between the signal and WAP domain. However, the WAP domains, even though variable, displayed certain conserved patterns. Generally, in WAP proteins, the WAP domain is marked by the beginning of KXGXCP motif. However, MC-crustin has HXGXCP motif instead. A Proline residue was found to be conserved adjacent to C1 residue of WAP domain. The location of aspartic acid and lysine residues across all crustins and WAP domain containing proteins was highly conserved, following the pattern of C3XXDXXC4XXXXKC5C6. Another observation was that among the crustins, adjacent to the second cysteine residue, no methionine residue was observed as was seen for the WAP domains of Elafin and SLPI (Fig. 6).

Fig. 6
figure 6

MUSCLE alignment of crustin peptides using default parameters of the Jalview version 2 workbench. MC-crustin was aligned against crustin sequences of crustaceans from the NCBI database. The WAP domain of Elafin and SLPI was also included in the alignment. Percentage identity colour scheme of Jalview was applied in the alignment

Full size image

Phylogenetic analysis revealed that Type I, Type II and Type V crustins are forming a monophyletic group and the Type III and Type IV crustins another group. MC-crustin was well nested within the Type I crustins and was grouped with the Scylla crustins. The Type I crustins were observed to have a common ancestor that included the carcinins (from Carcinus maenas). The monophyletic group consisting of carcinins were observed to contain Type I crustin sequences from crabs, lobsters and crayfish (Fig. 7).

Fig. 7
figure 7

Evolutionary tree of Crustins inferred using the Neighbor-Joining method. Percentage of replicate trees in which the associated taxa clustered together in the bootstrap test (1000 replicates) are shown next to the branches. All ambiguous positions were removed for each sequence pair (pairwise deletion option)

Full size image

Secondary structure analysis

Secondary structure prediction by RaptorX for MC-crustin showed coil regions predominating (98%) and presence of strand (1%) within the sequence (Fig. 8). TANGO did not predict any beta aggregation regions. However, AGGRESCAN predicted aggregation prone segments for the peptide (Fig. 9). Similarly, amyloidogenic regions of 21–25 and 73–77 were predicted by amyloid tool (Fig. 10).

Fig. 8
figure 8

Distribution of the three state secondary structure class of MC-Crustin predicted by RaptorX

Full size image
Fig. 9
figure 9

a Graph demonstrating the aggregation tendency profile of the sequence in the MC-crustin mature peptide. b Graph portraying exclusively the area employed in a hot spot (normalized by the peptide’s length). Regions 23–26 and 84–90 of the mature peptide are foreseen as potential aggregation regions

Full size image
Fig. 10
figure 10

FoldAmyloid predicted amyloidogenic regions in MC-crustin: A21L22Y23C24C25 and C73C74Y75D76A77

Full size image

Tertiary structure analysis

SWISS-MODEL user template-based prediction provided a 3D structure for the WAP domain of MC-crustin which was modelled based on Elafin [PDB ID:1FLE (chain B)] (Fig. 11a). Pdbsum analysis revealed the structure with coil, helix and strand conformations (Fig. 11b). Ramachandran plot analysis showed more than 90% residues within the most favoured and additional allowed regions (Fig. 11c). Even though the mature peptide full length was provided as sequence input, the output predicted structure was only available for the WAP domain region of Histidine 39 (H39)–Valine 87 (V87). PDBsum structural details revealed that four strands formed the polypeptide chain fold with external two strands (strand 1 and strand 2) and internal core strands (β1 and β2) forming beta-sheets. Strand 1 is connected to strand 2 via a larger loop and strand 2 is connected to the internal beta-strand 1 (β1) by a helical region (D64, G65, A66) (Fig. 11d) and a γ-turn like segment (F69, R70, S71) (Fig. 11f) leading to the beta-strand 1. β1 then takes a γ-turn (D76, A77, C78) and a β-hairpin (D76, A77, C78, V79, E80, H81) (Fig. 11e) to position the second beta strand forming the anti-parallel beta sheet. The overall structure is further stabilized by the disulphide bonds of cysteines C1–C6, C2–C7 and C3–C5 connecting all the four strands with one another and the fourth disulphide bond C4–C8 that connects the first and second loops. The disulphide bonds, distribution of cationic and hydrophobic amino acid residues and the electrostatic potential distribution were predicted by Swiss-PDB viewer (Figs. 11 and 12). The loop region prior to WAP domain was observed to have a greater positive charge distribution (Fig. 12b).

Fig. 11
figure 11

a Tertiary structure of WAP domain of MC-crustin based on Elafin chain B (1FLE) template predicted by SWISS-model. The structure was viewed and disulphide bonds were predicted in Swiss-Pdb viewer version 4. b Ramachandran plot for predicted structure. c Wiring diagram predicted by PDBsum showing MC-crustin’s secondary structural elements (alpha-helices and beta-sheets) together with various structural motifs such as d helices, e beta-hairpins and f gamma-turns

Full size image
Fig. 12
figure 12

a Swiss-PDB viewer prediction of the cationic (blue) and hydrophobic residues (green). b Electrostatic distribution denoting positive charged regions (blue) and the negatively charged regions (red)

Full size image

Therapeutic potential

The therapeutic potentials of MC-crustin were predicted using several online prediction tools and the prediction scores above the threshold were considered as positive and below as negative (Table 2). MC-crustin was predicted to be a bioactive peptide by Peptide Ranker with a score of 0.986. iAMP was employed to predict three classes of AMPs as anti-bacterial, anti-fungal and anti-viral. MC-crustin was predicted to be anti-bacterial, anti-fungal as well as anti-viral. The anti-viral property was further predicted by FIRM-AVP to be positive. Pro-inflammatory peptides are a class of peptides that can be used as therapeutic candidates to treat and alleviate several disorders. PIP-EL predicted MC-crustin to have pro-inflammatory property. Chronic inflammation and auto-immune diseases develop when inflammatory responses continue after they have completed their normal function. MC-crustin was predicted as anti-inflammatory by Pre-AIP as well as AIP-Pred tools. PredAPP predicted the peptide to have anti-parasitic potentials with a prediction potential of 0.62. Anti-angiogenic peptides regulate angiogenesis and tumour growth. MC-crustin was predicted to have anti-angiogenic property with a prediction value of 1.48. The peptide was also predicted with anti-hypertensive property by the AntiangioPred tool, and the WAP domain seemed to have higher prediction score of 0.79. The peptide was predicted as non- cell penetrating type and non-haemolytic.

Table 2 Therapeutic potentials of MC-crustin predicted in silico
Full size table

Discussion

For decades, the unregulated use of antibiotics for therapeutic purposes and as prophylactics in agriculture and animal husbandry has led to the emergence of antimicrobial resistance (AMR) in an alarming rate across the globe. Search for alternatives to antibiotics are on the run to tackle this crisis (Iskandar et al. 2022). In this context, the potentials of naturally evolved antimicrobial peptides, present in all groups of organisms, right from bacteria to the plants, humans, etc. are being explored widely as lead molecules for management of various diseases (Erdem Büyükkiraz and Kesmen 2022). Marine organisms are exposed to a plethora of pathogens and hence have an efficient first line of defence against the invading organisms (Wu et al. 2021). Marine invertebrates defend themselves against pathogens mainly through innate immunity. With respect to crustacean immunity, there is a well-played orchestra of immunity factors such as agglutination, prophenoloxidase system, encapsulation, melanization and production of antimicrobial peptides to deal with the invaders (Cerenius et al. 2010).

Crustin in the present study (MC-crustin) was obtained from the gill transcripts of Scylla serrata, and was predicted to have 10.164 kDa with a theoretical pI of 8.27. In general, crustins range from 6 to 22 kDa, with pI values in the range 4–8.5 (Smith et al. 2008). A medium value of predicted Boman index denotes that MC-crustin may not be involved in protein binding like the receptor binding activity (Boman 2003). The negative GRAVY value denotes the peptide to have relatively lower hydrophobicity. This is further supported by the lower hydrophobicity ratio for MC-crustin. Charge of the peptide is positive with R and K contributing to the greater positive charge. Thus, the cationicity and electrostatic potential may be a driving force compared to the hydrophobicity for the peptide’s initial interaction with bacterial membranes (Moravej et al. 2018; López Cascales et al. 2018). The primary contact of peptides with pathogenic surfaces involves biochemical or biophysical affinity carried out via hydrophobic or electrostatic interactions (Moravej et al. 2018). Higher proline composition (11%) of MC-crustin is a notable feature as the importance of proline residues in antimicrobial activities have been well documented. Introduction of positive charges and proline residues in F5W-magainin resulted in increased therapeutic potential (Azuma et al. 2020). The proline residue in maculatin 1.1 facilitated membrane insertion via lipid disordering (Fernandez et al. 2013). A higher instability index of the MC-crustin mature peptide than the full-length peptide denotes the signal cleaved mature peptide as less stable than the precursor. The limited stability and short half-life of few hours of the active mature peptide may limit damage to the ‘self’, once their purpose of antimicrobial activity has been served (Brockton et al. 2007). The signal peptide cleavage site is predicted to be recognized by the Sec/SP1 system, with N-region, H-region and C-regions. The transmembrane helix region predicted from positions 5–27 of the full-length MC-crustin may be the region responsible for anchoring the nascent peptide to membrane of endoplasmic reticulum (Valore and Ganz 1992; Ganz 2003).

Classification and formula annotations for different types of crustins have been attempted by several authors, and it is still an area under perusal (Bartlett et al 2002; Smith et al 2008; Tassanakajon et al. 2015, 2011; Vargas-Albores and Martínez-Porchas 2017). Type1 crustins or the carcinin like peptides (Zhao and Wang 2008) have cysteine rich region, with four conserved cysteine residues between the signal peptide and WAP domain (Smith et al 2008; Matos and Rosa 2022). Crustin signature domain is characterized by 12 conserved cysteine residues among which the last eight cysteine residues are present in C-terminal WAP domain, also observed in MC-crustin (Ranganathan et al 1999; Bartlett et al. 2002; Smith et al. 2008). Other consensus patterns recognized in the C-terminal WAP domain for MC-crustin and others are: (i) HXGXCP containing C1, similar to the KXGXCP recognized as WAP motif (Ranganathan et al. 1999) and (ii) aspartic acid (D) and lysine (K) residues conserved as C3XXDXXC4XXXXKC5C6XDXC7 (Ranganathan et al. 1999; Bartlett et al. 2002; Smith et al. 2008). The region between C-terminal WAP domain and N-terminal signal peptide contains proline (P) and cysteine (C) rich region that follows the Type I crustin consensus pattern of C-X3-C-X (8–12)-C-C-X (16–17)-C-X6-C-X (9–10) (Smith et al. 2008). In mammals, the WAP domain is recognized to have serine protease inhibition, which is believed to be due to the presence of a methionine (M) residue adjacent to the second cysteine in the 4DSC (Ota et al. 2002; Francart et al. 1997). However, in WAP domain-containing proteins with antibacterial properties, this is replaced by a cationic or hydrophobic amino acid (Smith et al. 2008). MC-crustin and other crustins do not contain methionine residue at the respective site, indicating antibacterial activity.

Type II crustins hold a long glycine-rich hydrophobic region at the N-terminus followed by the 12 conserved cysteine containing ‘crustin signature domain’ which includes the C-terminal single WAP domain. The Type III/single WAP domain (SWD) crustins are defined by the presence/absence of a short proline and arginine-rich region at the N-terminus and a single WAP domain at the C-terminal of mature peptide. They do not contain the complete ‘crustin signature domain’ (Smith et al. 2008; Smith 2011; Tassanakajon et al. 2011). Type IV crustins/DWD crustins, as their name suggest, have two WAP domains (Smith 2011). Finally, Type V crustins from insect order Hymenoptera, are similar to Type I crustins; nevertheless, with an aromatic amino acid-rich domain between the cysteine-rich domain and WAP domain (Zhang and Zhu 2012). Zhao and Wang (2008) had proposed a divergent evolution for WAP domains in crustaceans, exhibiting different function such as antimicrobial activity and proteinase inhibitor activity (Du et al. 2010; Jiang et al. 2013; Zhang et al. 2017). Neighbour joining analysis in present study revealed two main branches of WAP domain containing proteins in crustaceans, one group containing the ‘crustin signature domain’ with WAP domain peptides (Type I and Type II crustins) and the other group which do not possess the signature crustin domain, yet containing the WAP domains (SWDs and DWDs), similar to previous studies such as Zhao and Wang, (2008) and Mu et al., (2010). In the first group of crustins, Type I and Type II crustins formed two distinct clades. Similarly, in second group, Type II and Type III also formed two distinct clades.

WAP domain structural analysis revealed antiparallel beta-sheet structure in the core region besides the major loop regions, stabilized by disulphide bonds (Ranganathan et al. 1999; Koizumi et al. 2008). Recent MD simulation studies in a crustin (Aacrus1) from the barnacle, Amphibalanus amphitrite revealed that the loop region is formed by 17 amino acid residues, extending from a serine (S) to phenyl alanine (F), adjacent to the first and second cysteine residues of its WAP domain, responsible for the deep anchoring of the peptide into bacterial membrane (Zhang et al. 2022a, b). The corresponding region in MC-crustin is from R37 to V53 of the mature peptide WAP domain region. However, the residues specifically involved in the binding needs to be assessed. Nevertheless, the loop region of MC-crustin from H39 to R54 was observed to have greater cationicity in the electrostatic potential distribution. It is likely that this region may be involved in strong binding to the negatively charged bacterial membranes and the stabilization of the structure due to the C1–C6 cysteine bond could be responsible for a stable binding and penetration into bacterial membranes, as observed in Aacrus1.

It is well known that AMPs show less ‘in solution’ aggregation and higher aggregation tendencies when in hydrophobic environments like bacterial membranes. Hydrophobicity influences aggregation of peptides. Increasing hydrophobicity in peptides beyond a critical level resulted in larger aggregates of peptides either ‘in solution’ itself or on bacterial membranes, consequently reducing its bioavailability. Hence, higher hydrophobicity and ‘in solution’ aggregation is generally considered to reduce the activity of the peptide. Nevertheless, optimum aggregation on membrane is essential for antimicrobial action (Torrent et al. 2011; Yin et al. 2012; Clark et al. 2021). MC- crustin did not show in solution aggregation (TANGO prediction) and only fewer aggregation prone residues as per AGGRESCAN aggregation tendency profile. The predicted aggregation hotspot regions (Y23–G26 and C84–Y90) and amyloidogenic regions (A21–C25 and C73–A77) confirm probability for peptide aggregations in bacterial membranes to execute their membrane disruption mechanisms (Yadav 2021; Kurpe et al. 2020; Zhang et al. 2021).

The potentials of MC-crustin were further supported by various in silico potential predictors, classifying the peptide as antimicrobial peptide. Search for biologically active peptide is increasing on account of the increased resistance by pathogens against the antibiotics in use. Bioactive peptides can be classified into short (< 20 aa) and long (> 20 aa) peptides (Mooney et al. 2012). In general, leucine (L), glycine (G) and lysine (K) are preferred for long bioactive antimicrobial peptides. Moreover, long bioactive peptides have certain general distribution of amino acid composition when compared to short bioactive peptides (3.9% of F, 7.5% of G, 5.8% of E, 4.8% of T, 4.5% of N, 5% of R). Ninety aminoacid sized MC-crustin was classified by Peptide Ranker as a bioactive peptide, falling in the long bioactive peptide class (Mooney et al. 2012). Antibacterial peptides have greater distribution of positively charged residues at the N and C-terminus of the peptides. Moreover, it has been observed that residues W, I, L, F are more frequent at the 2nd position of N-terminus of antibacterial peptides in comparison to non-antibacterial peptides. Residues such as G, R, K, L and C are in greater proportion in antibacterial peptides (Lata et al. 2007). MC-crustin had 7% each of G, R and K, 13% C and 6% G and greater distribution of positively charged amino acid residues—K, R and H in the C-terminus WAP domain, making it a suitable candidate as antibacterial peptide supplemented by prediction of iAMPpred (Meheret al. 2017). It has been postulated that higher frequency of the positively charged residues at C-terminus may help to interact with the negatively charged bacterial membranes (Lata et al. 2007).

Instability index greater than 40, slightly negative GRAVY score and abundance of aliphatic residues like I, V and L (total 15% for MC-crustin) are properties associated to be observed with antiviral peptides (Chang and Yang 2013). Crustins are known to show antiviral properties (Du et al. 2019; Wang et al. 2021a, b). iAMP pred supported the anti-viral potential for MC-crustin in the present study. Residues like C, G, H, K, R, and S are prominent in antifungal peptides. MC-crustin being a cysteine rich peptide (13% C) and equal distribution of lysine and arginine (7% each) as well as histidine and serine (6% each) could be the possible amino acid composition features suitable for antifungal peptide, supported by prediction from iAMPpred.

Inflammation is a biological response of the immune system that can be triggered by a variety of factors, including pathogens, damaged cells and toxic compounds (Chen et al. 2017). Proinflammatory peptides (or PIPs) are therapeutic candidates to cure various diseases that help in vaccine development. In general, arginine and leucine are preferred amino acids in proinflammatory peptides. The most abundant dipeptide compositions in PIPs are aliphatic-aliphatic, positively charged-aliphatic and hydroxyl group aliphatic or aromatic amino acids (FF, SL, SR, SF, SV, LL, LI, RT, RA) (Manavalan et al. 2018a, b). MC-crustin had equal distribution (1.12% each) of the dipeptide compositions SL, SF and LI in the sequence. Inflammatory responses are tightly controlled under normal conditions and are essential for the initiation of protective immunity. However, when they are not balanced after their routine functions, it can result in auto-immune disorders and chronic inflammation. For in silico predictions, arginine, leucine and lysine are observed to be dominant in anti-inflammatory peptides and the preferred dipeptide compositions are LL, SL, LE, LI, LS, LK, YL, IK and RI (Manavalan et al. 2018a, b). MC-crustin had equal distribution of SL, LI and YL dipeptide compositions (1.2% each). Moreover, the expression of crustin genes is reported to be mainly regulated by NF-κB transcription factors activated by both Toll and IMD signalling pathways, which are also a part of inflammatory signalling (Wang et al. 2013; Huang et al. 2016).

Angiogenesis promotes blood flow to injured tissues for healing. However, it is kept under regulation by several endogenous anti-angiogenic factors like anti-angiogenic peptides, thereby preventing tumour genesis (Rosca et al 2011). In MC-crustin, amino acids viz., C, P, S, R and W contributed to the anti-angiogenic property. MC-crustin was abundant with cysteines (13%) and prolines (11%), followed by arginine (7%) and serine (6%). Moreover, the N-terminal had S, P, W and C residues, a pattern observed in anti-angiogenic peptides (Ettayapuram Ramaprasad et al. 2015).

Anti-hypertensive peptides are known to target angiotensin-1 converting enzyme (ACE). Generally, anti-hypertensive peptides, which are inert inside the original protein, which become active once cleaved, have been reported to contain usually 2–20 amino acids (Himaya et al. 2012). All classes of anti-hypertensive peptides are known to have proline abundance, which was also observed for MC-crustin. Moreover, residues like aspartic acid and serine are known to be less frequent in these peptides (6% each in MC-crustin) (Kumar et al. 2015). A notable prediction is the WAP domain as anti-hypertensive, when compared to other regions of the mature peptide. Furthermore, the higher proline residue and reduced hydrophobicity may be responsible for the non-hemolytic property of the MC-crustin.

Conclusion

Marine derived AMPs have potential biomedical importance, making them anticipated alternatives for rational drug strategies and pharmaceuticals with a large potential in biomedical and aquaculture applications. MC-crustin with its cationic, cysteine and proline rich structure, was predicted with antimicrobial, anti-viral, anti-inflammatory, anti-angiogenic and anti-hypertensive property. Presence of aggregation prone segments with positive charge on WAP domain loop region are supportive of membrane binding and disruption property of the peptide. Being non-haemolytic with notable bioactive property, MC-crustin would be a promising drug molecule for therapeutic applications as an alternative to the antibiotics.