Marine-Derived Peptides from Phaeodactylum tricornutum as Potential SARS-CoV-2 Mpro Inhibitors: An In Silico Approach

The ongoing threat of viral pandemics such as COVID-19 highlights the urgent need for novel antiviral therapeutics targeting conserved viral proteins. In this study, peptides of 10-30 kDa derived from the marine diatom <i>Phaeodactylum tricornutum</i> were identified as potential inhibitors of SARS-CoV-2 main protease (Mpro), a key enzyme in viral replication. Peptides less than 60 amino acids in length were retrieved from the UniProt database and aligned with reference antiviral sequences using the Biopython pairwise2 algorithm. Six candidates were selected for structural modeling using AlphaFold2 and Swiss-Model, followed by molecular docking using ClusPro2. LigPlot+ was used to assess molecular interactions, while NetMHCpan 4.1 and AVPpred evaluated immunogenicity and antiviral potential, respectively. Molecular dynamics simulations over 100 ns were conducted using OpenMM. These peptides demonstrated stable binding interactions with key catalytic residues of Mpro. Specifically, peptide A0A8J9SA87 interacted with Cys145 and Glu166, while peptide A0A8J9SDW0 exhibited interactions with His41 and Phe140, both of which are known to be essential for Mpro inhibition. Although peptide A0A8J9X3P8 also interacted with catalytic residues, it exhibited greater structural fluctuations during molecular dynamics simulations and achieved lower AVPpred scores, suggesting lower overall antiviral potential. Therefore, A0A8J9SA87 and A0A8J9SDW0 were identified as the most promising candidates. Molecular dynamics simulations further supported the high structural stability of these peptide-Mpro complexes over a 100 ns timescale, reinforcing their potential as effective inhibitors. These findings support <i>P. tricornutum</i> as a valuable source of antiviral peptides and demonstrate the feasibility of <i>in silico</i> pipelines for identifying therapeutic candidates against SARS-CoV-2.