Recombinant Production and Molecular Docking Studies of Casoplatelin, a Bioactive Peptide

2.1. Chemicals, Materials and Reagents

Two single-strand oligonucleotides were procured from Microsynth, Switzerland. The Sequences are 5’ATGGCGATT CCGCCGAAAAAAAACCAGGATAAA3’ and 5’TTTATC CTGGTTTTTTTTCGGCGGAATCGCCAT3’. T4 DNA ligase was obtained from New England Biolabs, USA. The pBADgIII version C and E.coli TOP 10 were procured from Invitrogen, USA. Kits for plasmid and BAP purification and labeling were purchased from Qiagen (USA) and Ambion Inc. (USA), respectively. Ampicillin, Streptomycin and Arabinose were procured from Sigma, USA.

2.2. Synthesis and Annealing of Oligonucleotides

Two single-strand oligonucleotides were suspended in MilliPore water (Model:Elix 3 UV Water Purification System (120 V / 60 Hz), Millipore, USA) at a final concentration of 100 μM. Oligonucleotide suspensions were then stored at 95°C for 5 min and were allowed to cool down gradually for proper annealing. Annealed products were then separated using a non-denaturing polyacrylamide gel.

2.3. Cloning and Characterization of Cloned Oligonucleotide

Cloning and characterization of casoplatelin were performed by annealing the oligonucleotide into a pBADgIII vector after digestion with Hind III and Pvu II, followed by purification using a Qiagen QIA Quick purification column. The linearized vector and synthetic construct were ligated using T4 DNA ligase (New England Biolabs, USA). After inactivation of T4 DNA ligase at 65°C for 10 min, mung bean nuclease digestion was performed to remove nucleotide overhangs. The partially ligated vector was further purified using a Qiagen QIA Quick purification column, and ligation was then performed at the Pvu II site. Completely ligated vectors were then transformed into E.coli TOP 10 strain and cells were then plated on LB medium containing ampicillin (25 μg/ml). Colonies were then patched onto LB medium containing ampicillin (25 μg/ml) and streptomycin (20 μg/ml). After growth, plasmids were purified and characterized using a Dot-Blot Southern hybridization method with acrylamide gels from Denaturing Gradient Gel Electrophoresis (DGGE). After electrophoresis, gels were cut to fit the gel transfer stack and were subjected to a 5-min transfer step using the transfer membrane (Invitrogen, USA). Transfer membrane was then removed from the stack and was UV cross-linked for 10 min before hybridization [14]. Labeling was performed with BrightStar TM Psoralen-Biotin kits (Ambion Inc.) as described by the manufacturer.

2.4. Overexpression and Purification of Casoplatelin

Transformed clones were verified using Dot-Blot Southern hybridization and were then grown in the presence of arabinose (0.1%) to induce overexpression of the peptide (Invitrogen pBAD instruction manual https://tools.thermofisher.com /content/sfs/manuals/pbadgiii_man.pdf). The overexpressed peptide was then purified using Qiagen Ni-NTA purification kits (Qiagen, USA). The molecular mass of the purified pro-peptide was estimated using SDS-polyacrylamide gel electrophoresis in 20% Tris-tricine [15].

2.5. Platelet Aggregation Assays

The pro-peptide was prepared for assay by digesting it with pepsin in a gastric buffer. The anti-thrombotic activity was then verified through spectrophotometric analysis using a spectrophotometer (Model: UV1800, Shimadzu, Japan). Briefly, platelets for the assay were collected from 10-ml human blood samples in anticoagulant buffer containing 318 mM citric acid, 62 mM trisodium citrate, and 133 mM glucose, which were prepared in the lab. Platelets were then suspended in 200 μl of Tyrode’s buffer, prepared in the lab. Details of the reaction mixture are presented in Table 1.

Control reaction mixtures were prepared identically except that pepsin-digested fragments were omitted. Absorbance values were recorded at 500 nm and were plotted as a function of time. The overall methodological approach utilized for recombinant casoplatelin is depicted in Fig. (1).

2.6. Prediction of 3D Structure and Computational Model of Casoplatelin and its Interaction with Fibrinogen

The 3D structure of the peptide was generated to develop a computational model of casoplatelin and its interactions with fibrinogen using the ab-initio method (http://robetta. bakerlab.org) Robetta [16]. Robetta performs structure predictions by parsing the submitted sequence into putative domains and generates five structure models using comparative modeling or de novo structure predictions. These methods utilize hierarchies from BLAST, PSI-BLAST, FFAS03, or 3D-Jury to find protein homologs of known structure. When no positive matches are found, the sequence is used as a template for comparative modeling, and after further failure to find a positive match, de novo Rosetta fragment insertion is used to predict the structure.

Modeled structures were docked with fibrinogen using AutoDock 4.2 [17], which includes a two-step process in which the receptor is defined using a three-dimensional grid and then the peptide is docked using a Lamarckian genetic algorithm for pre-calculated grids. Agridbox of 126 × 126 × 126 was created with a spacing of 1 Å and a grid center of −95.332, 13.397, and −15.749 was used to calculate grids. Element maps that were generated by AutoGrid for calculations included Ca²⁺ with other atom-specific affinity maps (C, HD, N, O, and S). AutoGrid also generates electrostatic and desolvation potential maps. Five peptide models from Robetta were docked onto fibrinogen using the Lamarckian Genetic Algorithm. In these computations,10 runs were performed with a population size of 150, a maximum number of energy evaluations of 2500000, a maximum number of generations of 27000, a mutation rate of 0.02, and a crossover rate of 0.8. The top 10 clustered docking results were ranked according to the lower binding energy. The peptide conformation that was ranked first among the ten was considered in further analyses using PyMOL and LIGPLOT [18].

3.1. Synthesis and Annealing of Oligonucleotides

Annealing of oligonucleotides was performed in annealing buffer at 95°C for 5 min to promote the formation of hydrogen bonds between complementary single-stranded DNA molecules. Annealed products were further characterized using gel electrophoresis, and a thick band corresponding to 50 bp was observed, indicating successful annealing (Fig. 2a).

3.2. Cloning and Characterization of Cloned Oligonucleotides

To achieve directional cloning, Sac I and Hind III pre-cleaved sites were added to both ends of the pBADgIII vector sequence. Because mobility shifts of 50 bp are difficult to measure, a single-stranded probe was added to the construct. BLASTN (https://blast.ncbi.nlm.nih.gov/Blast.cgi) [19] searches of known vector databases and the E.coligenome were performed using the probe sequence to eliminate non-specific binding to any region of the vector. The present pBAD expression system released the expressed product outside the cell with the gIII signal sequence. The pro-peptide was found as a fusion protein with a 6 × histidine tag in the periplasm. The protein digestion tool (https://web.expasy.org /peptide_cutter/) [20] was checked to see which peptide digestion products were released by pepsin digestion. This prediction showed that the predominant products after digestion were casoplatelin and the 6× histidine tag. The prediction of peptide digestion capability of ExPASy has been showcased in the case of casein-based ACE-inhibitory peptide [21] and angiotensin I-converting enzyme inhibitory peptide [22]. Thus, the use of ExPASy in correctly identifying the peptides from a precursor protein has been successful.

The linearized pBADgIII vector was then cloned with the annealed oligonucleotide and was then used to transform E.coli TOP10 cells. After growth in ampicillin rich LB media plates, resistant colonies were selectively grown to isolate plasmid for characterization in Dot-Blot Southern hybridization assays using the upward capillary transfer method. This method enabled the detection of recombinant clones with a non-isotope labeled probe (Fig. 2b). Usage of Southern hybridization assays for the detection of recombinant clones has been also reported with the recombinant CHH-B1 [23] and (CHH-B2) [24]. Clones were then verified using Dot-Blot Southern hybridization and were induced in the presence of 0.1% arabinose for overexpression of casoplatelin.

3.3. Overexpression and Purification of Casoplatelin

Analyses of supernatants in 20% Tris-Tricine SDS-polyacrylamide gels revealed a band corresponding to a 5-kDa protein, which was absent in control and was the estimated molecular mass of the pro-peptide (data not shown). The expressed product was purified using a Ni-NTA column purification kit from Qiagen, USA. A band corresponding to 5 kDa was seen in the elute (Fig. 3a) and the purified product was digested using pepsin in a gastric buffer.

3.4. Platelet Aggregation Assay

Effects of casoplatelin on ADP-induced aggregation of platelets were tested at a molar ratio with human fibrinogen of 30:1. Following the addition of the 30-fold molar excess of the peptide to the reaction, aggregation was monitored by recording absorbance at 500 nm every min for 15 min. The control comprised the reaction mix without inducer (ADP) and absorbance was recorded every 10 s for 1 min. After the addition of ADP, readings were taken every 5 min for 1 h at the same wavelength. Absorbance values were plotted as a function of time (Fig. 3b), which decreased rapidly in control, indicating progressive aggregation of platelets. In contrast, the absorbance of test samples containing peptide increased for the first two min and gradually decreased after that, with little change over ten min. These observations are direct evidence that casoplatelin inhibits ADP-induced platelet aggregation. The progressive aggregation of platelets coupled with anti-platelet activity has also been reported with Cc₃-SPase [25] and sheep lactoferrin [26].

3.5. Prediction of 3D Structure and Computational Model of Casoplatelin and its Interaction with Fibrinogen

The undecapeptide (residues 106-116) casoplatelin comprises regions with hydrophobic residues (106-109) and hydrophilic residues (110-116). The N-terminal undecapeptide of cow κ-caseinoglycopeptide (residues 106-116 of κ-casein) reportedly has anti-platelet activity and inhibits platelet aggregation. In addition, two smaller tryptic peptides (residues 106-112 and 113-116) and one longer peptide (residues 103-111 of κ-casein, around the chymosin) of this undecapeptide have been shown to have moderate effects on platelet aggregation but no anti-platelet activity [27].

Computational analyses of casoplatelin were performed using Robetta [16], which requires a minimum of 27 amino acids. Thus, we added eight κ-casein precursor amino acids onto both sides of the casoplatelin peptide, and predicted the structure of the resulting 27-residue sequence using 2kon as the template, which was identified from Robetta’s inbuilt Ginzu domain prediction tool. To build a computational model of casoplatelin binding to fibrinogen (Fig. 4), five models were predicted by Robetta with a score of 0.38, indicating that the predicted structure was of medium-high quality. Although peptides can generally assume any three-dimensional shape, Robetta predicted two models with beta-strand pairing (models 1 and 4), indicating a high probability that casoplatelin is enriched with β-strands.

To develop a computational model of casoplatelin-fibrinogen interactions, we performed molecular docking using AutoDock 4.2 [17, 28]. AutoDock has been used previously to predict interactions between peptides and proteins, and molecular docking analyses have been successful in similar studies [29, 30]. Five models were prepared for AutoDock with 32 maximum torsion or rotatable bonds for each molecule. Although the peptide model contained more than 100 rotatable bonds, we reduced this number to 32 by making amide and backbone bonds as non-rotatable. Additionally, polar hydrogens and partial Gasteiger charges were added to complete the ligand processing step [31, 32]. The molecular docking results of casoplatelin showed interactions with fibrinogen A, B, or C chains in each of the peptide models as shown in Fig. (5).

Specifically, Model 1 interacted with the C chain, Model 2 interacted with A and C chains, Model 3 interacted with the B chain, and Model 5 interacted with the C chain. Fibrinogen predominantly comprises coiled-coil regions, and terminates with specific globular domains of domain D, where the C-terminal ends of Bβ and Bγ and Aα are interlinked (Table 2). Specifically, in the Aαchain, the C-terminal region extends from domain D, which can interact with another Aα C-terminal region, and also with the E domain. These types of intermolecular interactions with fibrinogen lead to clot formation via cross-linking. Thus, fibrinogen E and D domains contain crucial interaction regions for conversion to fibrin. The molecular interactions of a peptide with fibrinogen towards anti-plate activity are also reported through the molecular docking studies in the case of Mytilusedulis protein hydrolysates [33] and Staphylococcus aureus coagulase [34].