Computational and in vitro experimental analyses of the anti-COVID-19 potential of Mortaparib and Mortaparib Plus

COVID-19 pandemic caused by SARS-CoV-2 virus has become a global health emergency. Although new vaccines have been generated and being implicated, discovery and application of novel preventive and control measures are warranted. We aimed to identify compound/s that may possess the potential to either block the entry of virus to host cells or attenuate its replication upon infection. Using host cell surface receptor expression (Angiotensin-converting enzyme 2 (ACE2) and Transmembrane protease serine 2 (TMPRSS2) analysis as an assay, we earlier screened several synthetic and natural compounds and identified candidates that showed ability to downregulate their expression. Here, we report experimental and computational analyses of two small molecules, Mortaparib and Mortaparib Plus that were initially identified as dual novel inhibitors of mortalin and PARP-1, for their activity against SARS-CoV-2. In silico analyses showed that Mortaparib Plus , but not Mortaparib, stably binds into the catalytic pocket of TMPRSS2. In vitro analysis of control and treated cells revealed that Mortaparib Plus caused downregulation of ACE2 and TMPRSS2; Mortaparib did not show any effect. Furthermore, computational analysis on SARS-CoV-2 main protease (M pro ) that also predicted the inhibitory activity of Mortaparib Plus . However, cell based anti-virus drug screening assay showed 30~60% viral inhibition in cells treated with non-toxic doses of either Mortaparib Plus or Mortaparib. The data suggests that these two closely related compounds possess multimodal anti-COVID 19 activities. Whereas Mortaparib Plus works through direct interactions/effects on the host cell surface receptors (ACE2 and TMPRSS2) and the virus protein (M pro ), Mortaparib involves independent mechanisms, elucidation of which warrants further studies.


4
proteins implicated as potential targets, host cell membrane proteins-TMPRSS2 (Transmembrane protease serine 2) and ACE2 (angiotensin-converting enzyme 2) play a central role in viral infection and entry (5,18,19). The glycosylated spike protein of SARS-CoV-2 initiates the infection by binding to the cell surface receptor ACE2 of the host cell (5,18), following which it gets cleaved and activated by host cell membrane protein TMPRSS2 (5,19,20). After the attachment, fusion, and entry of the virus into the host cell, the genetic material of the virus (positive sense ssRNA) gets translated into various polyproteins using host cell translation machinery. However, many of the translated polypeptides exist in an inactive state and are converted to functional state upon cleavage by other SARS-CoV-2 proteases. Among these translated polyproteins, a crucial viral protease enzyme called M pro has been shown to activate itself and cleave other SARS-CoV-2 polyproteins (21)(22)(23) that are essential for virus replication, assembly, and transmission. SARS-CoV-2 has been reported to possess higher binding affinity to ACE2 as compared to SARS-CoV (24) suggesting ACE2 as a potential target. However, ACE2 receptor protein is involved in host cells' physiological functions, thus, its strong suppression could be counterproductive (25)(26)(27)(28)(29). TMPRSS2, on the other hand, has been shown to be involved in pathological processes (25,30). M pro is coded by the SARS-CoV-2 using the host cell resources and is pathologically highly relevant. In 5 Corporation, Osaka, Japan)-supplemented with 5% fetal bovine serum (Thermo Fisher Scientific, Japan) and 1% penicillin/streptomycin (Invitrogen, Carlsbad, CA, USA) in a humidified incubator with 5% CO 2 at 37°C. A library of 12,000 compounds (synthetic and natural) was earlier screened for anticancer candidates capable of abrogating mortalin-p53 interactions resulting in nuclear enrichment of p53 and a shift in mortalin staining from perinuclear (concentrated around the nuclear membrane; typical of cancer cells) to pancytoplasmic (widely distributed in the cell cytoplasm; typical of normal cells) type (31).
Out of 12,000 compounds screened, only 6 were finally selected and further tested for their anticancer potential in in vitro and in vivo analyses. Two out of 6 (CL-44, named Mortaparib; and CL-49, named Mortaparib Plus ) showed considerable anticancer potential as they inhibit mortalin and PARP1 (31,32). In the present study, binding affinity of CL-44/Mortaparib (5-

Preparation of protein and ligand structures
To analyze the effect of Mortaparib and Mortaparib Plus against the SARS-CoV-2 infection and multiplication, three crucial proteins were targeted, namely, M pro , TMPRSS2, and ACE2. The 3D structure of M pro and ACE2 was retrieved from protein data bank (PDB) having PDB ID 6LU7 (33)

Grid generation and molecular docking
After preparation of the protein and ligand structures, the grid was generated at the active site of all the proteins to dock both ligands. In case of M pro a grid of 20 Å edge was generated around Phe140, Asn142, Gly143, His164 and Glu166, as these residues make polar contacts with the already known peptide-like inhibitor N3 in the crystalized structure of M pro (PDB ID: 6LU7) (33). In the case of TMPRSS2, the grid was generated by choosing the three main catalytic residues namely, His296, Asp345, and Ser441 (40). While for ACE2, it was found that in the native crystal structure, Ser19, Gln24, Lys34, Glu35, Asp38, and Gln42 of ACE2 were involved in the hydrogen bonding with receptor binding domain of Spike protein (41). So, the grid was generated around these interacting residues. Subsequently, the extra precision flexible docking was performed for all the prepared systems using the Glide module of Schrodinger suite (42).

Molecular Dynamics (MD) simulations of the docked systems
The best-docked poses of the protein-ligand complexes were further taken for MD simulations using Desmond from the Schrodinger suite (36). The MD simulation protocol used here has been described in detail in our previous study (43)

Reverse Transcription Quantitative PCR (RT-qPCR)
A total of 2 x 10 5 cells per well were plated in a 6-well plate, allowed to settle expression levels, the geometric mean of the housekeeping gene 18S was used as an internal control. Details of the primers are listed in Table 1.

Western blotting
T.Tn cells (20 x 10 4 / well) were plated in a 6-well plate, allowed to settle overnight,

Antiviral activity assay
The assay was done in a 96-well plate in triplicates.

Statistical analysis
The mean and standard deviation of data from three or more experiments were calculated.
The degree of significance between the control and experimental samples was determined using an unpaired t-test (GraphPad Prism GraphPad Software, San Diego, CA, USA).

Computational analyses of Mortaparib and Mortaparib Plus as potential inhibitors of TMPRSS2 and ACE2
We showed hydrogen bonding with Ser436, and polar interactions with two main catalytic residues His296 and Ser441 ( Fig. 2A). In case of ACE2-Mortaparib Plus interactions, ACE2 residues (Lys31 and Glu75) that interact with viral Spike protein were found to be involved in hydrogen bonding. Glu35 also showed columbic interactions with Mortaparib Plus in the best docked pose (Fig. 2B). The docking score as well as binding characteristics (polar and non-polar interactions) for all the complexes are listed in Table 2. These docked complexes were also taken further for MD simulation of 100 ns to investigate the stability and dynamic behavior of the proteins when bound to Mortaparib Plus . When MD trajectories were examined, it was found that Mortaparib Plus stayed at the binding site of both the proteins (Fig.   2C). The bound ligand did not show fluctuations during the simulations, as shown in the RMSD plot (Fig. 2D). The trajectories of protein and ligand complexes were also analyzed for investigating deviations in the overall structure. The RMSD plot showed structural stability without any significant deviation for all the complexes (Fig. 3A). Similarly, in RMSF, no major fluctuation was seen throughout the duration of simulation (Fig. 3B).
Analysis of hydrogen bonds for each protein-ligand complex revealed that on an average  (Fig. 3C). When significant fraction of interaction (interaction fraction >30% of the simulation time) was calculated for the entire course of simulation, it was found that in case of TMPRSS2, two residues (Tyr459 and Tyr474) made significant water-based interactions with Mortaparib Plus , but none of the main catalytic residues (His196, Asp345 and Ser441) were involved in major interactions (Fig. 3D). Similarly, in case of ACE2, only one residue (Asn61) was involved in significant interaction, while none of the residues of ACE2 that make hydrogen bonds with the Spike protein (Ser19, Gln24, Lys34, Glu35, Asp38 and Gln42) significantly interacted with Mortaparib Plus (Fig. 3E) (Fig. 3F). These computational analyses suggested that (i) Mortaparib Plus , but not Mortaparib, could bind stably with ACE2 and TMPRSS2, and (ii) the interactions were significantly stronger with TMPRSS2.

TMPRSS2 and ACE2 expression analyses in cells treated with Mortaparib or Mortaparib Plus
In the MTT-based cell viability assay, Mortaparib and Mortaparib Plus showed dosedependent toxicity in the T.Tn cells. Mortaparib Plus was relatively stronger than Mortaparib (Fig. 4A). Similar cytotoxicity has been reported earlier in a variety of other cancer cell types (31,32). In the present study, we selected non-toxic dose (1 M) for both the compounds.
T.Tn cells were treated with 1 M (Mortaparib or Mortaparib Plus ) for 48 h followed by the expression analysis of the target proteins. As shown in Fig. 4B, Mortaparib Plus -treated cells

Mortaparib and Mortaparib Plus as a potential inhibitor of viral protein M pro
We Thr190 of M pro were involved in other polar and hydrophobic interactions (Fig. 5A). The docking score and details of interacting residues are listed in Table 2. The docked complex was then simulated in water for 100 ns to investigate the stability and dynamic behavior of the M pro -Mortaparib Plus complex. Mortaparib Plus was found to bind stably at the active site in M pro throughout the simulation (Fig. 5B). RMSD plot of Mortaparib Plus also showed its stability within the M pro pocket (Fig. 5C). The overall structure of the bound protein also did not show any significant deviations (Fig. 5D). Similarly, in RMSF analyses, no major fluctuations were seen (Fig. 5E). Next, the interaction fraction time of each active site residue with the Mortaparib Plus was calculated for the simulated trajectory, which showed that the three main catalytic and conserved residues (His41, Met165 and Gln192) of M pro made significant interactions (i.e., >30% of the simulation time) with Mortaparib Plus (Fig. 5F). The average number of hydrogen bonds that Mortaparib Plus made with M pro during the MD run was 0.99±0.29 (Fig. 5G). The characteristic values of Mortaparib Plus in terms of RMSD, RMSF, Radius of gyration and SASA are listed in Table 3. Finally, the MM/GBSA free energy of Mortaparib Plus binding to M pro showed a strong affinity (-47.37 ± 0.07 kcal/mol) ( Fig. 5H), which was much higher than that for the host cell proteins TMPRSS2 and ACE2.
This data indicated that Mortaparib Plus may possess the potential to inhibit the activity of M pro as well. Taken together, these in silico data suggested that Mortaparib Plus may inhibit virus infection by its impact on host cell receptors as well as virus proteins.

Discussion
Host cell surface receptor protein, TMPRSS2, is known to interact with the SARS-CoV-2 virus and cause splicing of the viral S protein into S1 and S2 (20). The former is essential for the virus-to-host fusion, while the latter interacts with the host cell receptor ACE2 and helps in internalization. Both TMPRSS2 and ACE2 are known to be enriched in lung, heart, kidney and intestinal endothelia, and operate as the first line of defense against the foreign pathogens. TMPRSS2 has been suggested to play a significant role in several physiological and pathological mechanisms involving internalization (such as the facilitation of sperm function) or inflammatory response (such as airway defense) (25)(26)(27). It has been shown to be pathologically upregulated in the prostate and colon cancer cells (40,45). ACE2, on the other hand, is commonly known to physiologically catalyze the hydrolysis of angiotensin II to angiotensin and contribute to vasodilatation (28,29), and to pathologically facilitate SARS-CoV-2 and similar infections (46). Thus, while the TMPRSS2 and ACE2 proteins are some of the major targets of the coronaviruses, their excessive suppression by targeted therapies may result into the appearance of unprecedented collateral adverse effects such as the male infertility, frequent respiratory infections, and hypertension (47,48).
Another crucial target to fight against SARS-CoV-2 is M pro , which cleaves the polyprotein into functional proteins to help in its replication process (49). was predicted to block TMPRSS2 more strongly than ACE-2. Taken together with the dispensable role of TMPRSS2, Mortaparib Plus is predicted to be a safer candidate therapeutic alternative for SARS-CoV-2 (50-52). Furthermore, it was also predicted to inhibit viral M pro that is crucial for viral replication in the host cells ( Fig. 4 and Tables 2-3