SARS-CoV-2 envelope protein attain K_ac mediated dynamical interaction network to adopt ‘histone mimic’ at BRD4 interface

Abstract

Interface mimicry, achieved by recognition of host-pathogen interactions, is the basis by which pathogen proteins can hijack the host machinery. The envelope (E) protein of SARS-CoV-2 is reported to mimic the histones at the BRD4 surface via establishing the structural mimicry; however, the underlying mechanism of E protein mimicking the histones is still elusive. To explore the mimics at dynamic and structural residual network level an extensive docking, and MD simulations were carried out in a comparative manner between complexes of H3-, H4-, E-, and apo-BRD4. We identified that E peptide is able to attain an ‘interaction network mimicry’, as its acetylated lysine (Kac) achieves orientation and residual fingerprint similar to histones, including water-mediated interactions for both the Kac positions. We identified Y59 of E, playing an anchor role to escort lysine positioning inside the binding site. Furthermore, the binding site analysis confirms that E peptide needs a higher volume, similar to the H4-BRD4 where both the lysine’s (Kac5 and Kac8) can accommodate nicely, however, the position of Kac8 is mimicked by two additional water molecules other than four water-mediated bridging’s, strengthening the possibility that E peptide could hijack host BRD4 surface. These molecular insights seem pivotal for mechanistic understanding and BRD4-specific therapeutic intervention.

KEY POINTS

1. Molecular mimicry is reported in hijacking and then outcompeting the host counterparts so that pathogens can rewire their cellular function by overcoming the host defense mechanism.

2. The molecular recognition process is the basis of molecular mimicry. The E peptide of SARS-CoV-2 is reported to mimic host histone at the BRD4 surface by utilizing its C-terminally placed acetylated lysine (K_ac63) to mimic the N-terminally placed acetylated lysine K_ac5GGK_ac8 histone (H4) by interaction network mimicry identified through microsecond molecular dynamics (MD) simulations and post-processing extensive analysis.

3. There are two steps to mimic: firstly, tyrosine residues help E to anchor at the BRD4 surface to position K_ac and increase the volume of the pocket. Secondary, after positioning of K_ac, a common durable interaction network N140:K_ac5; K_ac5:W1; W1:Y97; W1:W2; W2:W3; W3:W4; W4:P82 is established between K_ac5, with key residues P82, Y97, N140, and four water molecules through water mediate bridge. Furthermore, the second acetylated lysine K_ac8 position and its interaction as polar contact with K_ac5 were also mimicked by E peptide through interaction network P82:W5; W5:K_ac63; W5:W6; W6:K_ac63.

4. The binding event at BRD4/BD1 seems an induced-fit mechanism as a bigger binding site volume was identified at H4-BRD4 on which E peptide attains its better stability than H3-BRD4.

5. We identified the tyrosine residue Y59 of E that acts like an anchor on the BRD4 surface to position K_ac inside the pocket and attain the interaction network by using aromatic residues of the BRD4 surface.

Communicated by Ramaswamy H. Sarma

Keywords:

• SARS-CoV-2
• molecular mimicry
• interaction mimicry
• protein–protein interaction
• molecular dynamics simulation
• envelope protein
• BRD4/BD1
• K_ac

5. Availability of data and material

We have used open-source tools and software to produce the computational data available and free to use for academic purposes. The PDB database (https://www.rcsb.org/) was utilized to obtain the crystal structures. For protein-protein docking, the software and servers used are HDOCK, HPEPDOCK, SwarmDock, MDockPep, and pyDock. The licensed version of AMBER 16 is used for molecular dynamics simulation; however, apart from AMBER, the GROMACS 5.1.6 is also used for post-processing molecular dynamics simulations dynamics and energetics analysis. The binding site analysis was carried out using HOLE and POVME programs. The python script was built with the help of the website (https://seaborn.pydata.org/generated/seaborn.violinplot.html) to generate the violin plot for volume analysis. The PCA was generated using Bio3D. The VMD and XMGRACE tools are used for visual analysis and preparation.

Acknowledgments

The authors would like to acknowledge DBT and THSTI for providing support, research opportunities, and computational facilities. A.K.A is thankful to MK Bhan-Young Researcher Fellowship Program (HRD-12/4/2020-AFS-DBT).

Authors’ contributions

S. Asthana conceptualized the study. A.K. Agrahari and S. Asthana established the rationale of the study. A.K. Agrahari curated and generated the data. A.K. Agrahari and S. Asthana did the analysis. M. Srivastava and M. Singh helped in the data analysis. A.K. Agrahari and S. Asthana wrote the manuscript. All authors have read and approved the manuscript.

Disclosure statement

There are no potential conflicts of interest.

Additional information

Funding

The author(s) reported there is no funding associated with the work featured in this article.