RELATED APPLICATION

This application takes priority from Indian Application 202041022202 filed 27/8/2020 and 202141026347 filed 14/6/2021 and are incorporated herein in its entirety.

FIELD OF THE INVENTION

The present invention is related to 3D bioprinted models for COVID-19 that simulates endothelial dysfunction, viral entry and other associated inflammatory responses caused by the virus and thus can be used for discovering new drugs , repurposing existing drugs used for other indication, testing monoclonal and polyclonal antibodies/passive vaccines with neutralizing properties and other nutraceuticals, herbal products for COVID-19. Using the platform technology of 3D bioprinting applicants have identified Ertugliflozin as a novel candidate for immediate repurposing as a treatment option for COVID-19 infections.

BACKGROUND OF THE INVENTION

Mitigation of COVID-19 prophylactical ly and therapeutically has been proposed and are currently in practise. In early 2020, many anti-viral compounds were proposed (WHO Solidarity trial results, 2021), natural products were reported (Natural and Nature-Derived Products Targeting Human Coronaviruses. 2021, Molecules, 26(2), 448) and now, vaccines were being developed (SARS-CoV-2 vaccines strategies: a comprehensive review of phase 3 candidates. 2021, npj Vaccines, 6(1), 1-17). In silico studies on such mitigation were also proposed by various groups. Universally, it is agreed that drug-repurposing is one of the safest and best bet towards the mitigation of fast prevailing COVID-19 for alleviating symptoms.

Drug repurposing refers to the identification of novel applications for an approved or investigational/experimental drug outside the premise of its medical indication. At present, this strategy would be a logical choice for developing a new drug for COVID-19 considering the substantial time-scales associated with new drug discovery and the trial-based validation of its safety and efficacy. The major advantage of repurposed drug is that it has been already evaluated for safety in human trials, which would reduce the significant amounts of time and money, a priority concern in SARS-CoV-2 drug development.

In the recent times, drug repurposing approaches have gained importance as these methods are faster and require less economic investment. Application of in silico techniques in early stages of drug discovery either through conventional or repurposing approaches aid in minimizing the chances of the failures. The SARS-CoV-2 virus is closely related to the SARS-CoV and this allows utilization of the known protein structures to quickly build a model for drug discovery to search for possible medications for the SARS-CoV-2. The aim of the present study is to in silico identify clinically approved drugs treating various diseases of human importance, which would be pharmacologically evaluated for targeting the RBD domain of the S-protein of COVID-19. The RBD domain of S-protein is an attractive and well characterised drug target in corona viruses owing to the pivotal role it plays in the entry of the virus into the host cell.

Targeting the above mentioned specific structural domain with repurposed drugs to disrupt the COVID-19 viral attachment to the host proteins is disclosed in Provisional Application 202041022202 incorporated herein in its entirety. Much research has gone into finding drugs that can effectively prevent or reduce viral-ACE2 interactions. Subsequently, testing their binding efficacy in vitro assays with the RBD will elucidate the computational rationale for robust proposal of drug repurposing. Screening the FDA approved drugs and antivirals from the Drugbank database through molecular docking will identify and prioritize repurposed drugs as potential leads. A virtual screening of 8000 FDA approved small molecule therapeutics including antiviral, anti-malarial, anti-parasitic, anti-fungal, anti-tuberculosis and active phytochemicals against RBD spike proteins of SARS-CoV-2 yield prioritized drugs for mitigation of COVID-19

The drugs identified by virtual screening were then tested in a cell-free biochemical RBD-ACE2 binding study to validate findings from virtual screening. In addition, the "DRUG-hits" from this biochemical screen were further profiled in cytokine secretion and three dimensional (3D)- printed human-cell-based vascular lung model for potential anti-inflammatory and anti-clotting activities which could be effective in treatment of COVID19.

Brief Description of Drawings

Scheme 1: 3D bioprinted vascular lung/kidney/heart model

Figure 1: Monomer S-Protein- Structural Classification

Figure 2: Active sites of RBD domain of SI protein.

Figure 3: Rank score and binding energies of selected drugs in RBD's site-1 & site-2.

Figure 4: Binding orientation of Homoharrigtone within the active site of the RBD site-1 and site-2.

Figure 5: Binding orientation of Ritonavir within the active site of the RBD's site-1 and site-2.

Figure 6: Binding orientation of Ertugliflozin within the active site of the RBD site-1 and site-2.

Figure 7: Binding orientation of Sitravatinib within the active site of the RBD site-1 and site-2.

Figure 8: Binding orientation of lodipamide within the active site of RBD site-1 and site-2.

Figure 9: Percent Inhibition by different drugs at different concentrations of 50pM, lOOpM and 200pM. *** P<0.001; ** P<0.01; *P<0.05. Student's t-tests were performed to compare between the percent inhibition shown by control and drugs.

Figure 10: IL-1 secretion by different drugs (all drugs at lOOpM).

*** P<0.001; ** P<0.01; *P<0.05. Student's t-tests were performed to compare IL-10 secretion between control and LPS treated cells; and between LPS treated cells (acting as a control) and different drugs treated cells.

Figure-11: Thrombodulin secretion by different drugs (all drugs at lOOpM).

*** P<0.001; ** P<0.01; *P<0.05. Student's t-tests were performed to compare thrombodulin secretion between control and LPS treated cells; and LPS treated cells (acting as a control) and different drugs treated cells

Figure-12: Effect on Monocyte adhesion by different drugs in 3D Vascular Lung Model.

*** P<0.001; ** P<0.01; *P<0.05. Student's t-tests were performed to compare between the control and drugs after treatment with LPS Detailed Description of the Invention

Respiratory tissue engineering has advanced within the last years and several approaches are described in the literature that can be used to mimic the human pulmonary epithelial tissue barrier by applying 2Dmodels ranging from simple mono-cultures to more sophisticated cocultures of various spheroids or organoids. A major drawback of the current 2D co-culture models is the use of commercially available two chamber systems, as well as culturing different cell types on two sides of a relatively thick (about 10 mm) porous membrane composed of materials such as for example poly-ethylene terephthalate (PET). This is not optimal for the development of especially a lung model where the real barrier corresponds merely to two very thin cytoplasmic lamellae of the endothelial and epithelial cells separated by a single basement membrane.

3D bioprinting represents an advanced, more physiological system since it involves cell-cell interactions mimicking the native tissue and its microenvironment. Furthermore since they are bioprinted using a printer there is accuracy of the structure and the cell architecture is much more representative of human tissues.

The present invention provides a 3D manufactured human like model to study damage caused by COVID-19 to lungs, kidneys and heart and is especially the need of the hour to find therapies for this pandemic. Using the same strategy, the present invention further provides Ertugliflozin as a candidate for immediate repurposing for treatment of COVID-19 infections.

3D BIOPRINTED MODEL OF THE INVENTION

The applicants have developed 3D bioprinted vascular lung, kidney and heart systems that mimic events seen in humans and screen therapies that can be developed to combat this infection. There are no validated animal models for this disease that capture the steps seen in human disease, so a human like model is needed to design therapies. Specifically the present model 3D system comprises human cells as provided in Scheme 1 that captures the following steps observed in patients:

1. The interaction of virus with lung capillary endothelium and induction of endothelial dysfunction; 2. The binding and entry of virus into lung epithelial cells;

3. The effect of virus on immune cell driven inflammation and cytokine secretion. This is measured by addition of human lymphocytes, monocytes and neutrophils to the 3D bioprinted vascular lung;

4. While the methods are described for a vascular lung model, it is noted that similar heart and kidney models can be created by replacing the lung airway epithelial cells by human HEK kidney cells for the kidney model and human cardiomyocyte cells for the heart model.

In one embodiment the 3D bioprinted platform provides for evaluation of the following: a) Effects on endothelial dysfunction and blood coagulation biomarkers induced by the virus b) Effect on binding of virus to endothelial and lung cell/kidney/heart layers in 3D bioprinted system c) Entry of virus into endothelial and lung/kidney/heart cells in 3D bioprinted system d) Effect of virus on Inflammatory response and cytokine secretion induced by virus

Shown in Scheme 1 below is the typical 3D bioprinted vascular lung/kidney/heart model of the invention. The basement membrane will be printed first followed by printed layering of either lung, kidney or heart cells. As an example, the examples included here show layering of lung cells. Another layer of collagen scaffold will be layered, and finally endothelial cells will be layered. Drugs can be added to the system to measure their effects on: a) Endothelial inflammatory monocyte adhesion and blood clotting biomarkers b) Damage to lung/kidney or heart cells in terms of cell death c) Associated inflammation and cytokine "storm" secretion created by virus

The invention is further enabled by the Examples provided herein. However, these Examples are for better understanding and may not be construed to limit the extent of the invention.

Examples

Example 1: Molecular Docking Studies Cresset Flare software was used for molecular docking studies against the spike protein SARS- CoV-2 (http://www.cresset-group.com/flare/).

Ligand preparation

The 2D structures of all the drugs were downloaded from Drugbank database and prepared using Flare software. Hydrogen atoms were added in the structure and atom force field parameterization was assigned. Further, energy minimization was done for all the drugs, nonpolar hydrogen atoms were merged, and rotatable bonds were defined. Later, ligand minimization was carried out in Flare by Minimize tool by using Normal calculation methods. Ligands should also be prepared to assign proper bond orders and to generate the correct tautomer and/or ionization state.

RBD structure preparation

The RBD of spike glycoprotein SARS-CoV-2 was used for present study. This RBD binds to ACE2 receptor on the host cell with high affinity, which makes it a key target for the novel coronavirus therapy development. The 3D structure of RBD binds to ACE2 receptor (PDB ID: 6M0J) were downloaded from Protein Data Bank (PDB) (https://www.rcsb.org). The protein has two chain A and E, the A chain has ACE2 receptor and E chain has RBD domain. The RBD domain was saved into PBD format for further studies. The target protein preparation was carried out in Flare software with default settings. Missing residues, hydrogens and 3D protonation were carried out on the target protein. Protein minimization has been carried out in Flare by Minimize tool by using Normal calculation methods.

Computational analysis of binding sites

Binding site was generated by Accelrys Discovery Studio visualizer 3.5 (Accelrys Software Inc.) to explore potential binding sites of the RBD protein using receptor cavities tools. Based on a grid search and "eraser" algorithm, the program defines where a binding site is. The binding sites were displayed as a set of points (point count) and the volume of each of cavity was calculated as the product of the number of site points and the cube of the grid spacing. Volume of each site was calculated and further saved and exported into Flare for advance analysis.

Database Search and Library Establishment Databases compound such as Drugbank were searched for screening the small molecules targeting the RBD of S- protein that contains key structural domains. Then, by using Lipinski's "rule of five", molecules with less reasonable physicochemical parameters were discarded, leading to the selection of candidates with good drug-like properties: molecular weight (MW) <600 KDa, number of hydrogen bond donors <5, number of hydrogen bond acceptors <10, loglO partition coefficient (logP) <5, Polar surface area (PSA) range from 50-150A2 and no more than one violation of the abovementioned criteria. A virtual library of approximately 3458 compounds extracted from Drugbank database was used for further screening.

Docking with Lead Finder

The full-atom model of RBD was prepared and docking of ligands to the prepared model of RBD in the active sites was performed using Flare (Lead Finder) software (http://www.cresset- group.com/lead-finder/) by default configuration setting. The energy grid box for ligand docking was set at the geometrical centre of the active site residues to span 10 A in each direction. Two active sites were identified and in each sites centre residues were picked to define the active site grid box. Lead Finder assumes that the protein is rigid and analyses the possible conformations of the ligand by rotating functional groups along each freely rotatable bond. For each ligand pose, Lead Finder determines values of the free energy of binding, the VS score, and the pose ranking score by using its three built-in scoring functions. Three different scoring functions are available from Lead Finder.

Rank Score: optimized to provide accurate prediction of 3D docked ligand poses.

AG: optimized to provide an accurate estimate of protein-ligand binding energy, on the assumption that the pose is correct.

VS: optimized to provide maximum efficiency in virtual screening experiments, with a maximum discrimination between active and inactive compounds in virtual screening experiments.

The database drugs were docked in the RBD domain of S-protein SARS-CoV-2 by using Cresset Flare Docking software with accurate and slow mode. The best poses were generated and visualized in pose viewer and 3D images stored in storyboard.

Structure of the S-protein The S-protein has two subunits SI and S2, RBD situated in the SI subunit binds to the cell receptor ACE2 in the region of aminopeptidase N. Full spike protein has 1,273 amino acids in length, and the amino terminus and most of the protein is predicted to be on the outside of the cell surface or the virus particles. A fragment that is located in the SI subunit and spans amino acids 319-541 has the RBD domain. SI protein has N-terminal domain (13-305 amino acid residues) and the RBD domain (amino acids 319-541), within the RBD domain there is one motif which makes complete contact with the receptor ACE2, was referred to as receptor-binding motif (RBM) (437-508 amino acid residues). The RBM region is tyrosine rich. Among the 14 residues of RBM that are in direct contact with ACE2, six are tyrosine, representing both the hydroxyl group and hydrophobic ring, as shown in Figure 1. S2 is responsible for membrane fusion and comprised of fusion peptide (FP), heptad repeat 1 (HR1), heptad repeat 2 (HR2), transmembrane domain (TM), and cytoplasmic domain fusion (CP), is responsible for mediating viral fusion and entry. For all COVID infections, S is further cleaved by host proteases at the so- called S2' site located immediately upstream of the fusion peptide.

Identification of active sites

Applicants' in silico strategy was to design and screen the new drug targeting the S-protein that contains key structural domains RBD which plays a pivotal role in viral-host attachment. One such avenue of research is to find new repurposed drug that can attach to residues at the site of binding of the RBD to the ACE2. In the current study applicants discover several potential binding sites for molecules that can occupy such druggable pockets so as to inhibit virus-ACE2 binding in vitro. The X-ray model of the RBD was used to identify 3 possibly druggable pockets where drugs might bind. The active site volume and binding surface area of three pocket is representation in Table 1. Site-1 has the volume of 143.7 and site-2 has the active site volume of 109.4, the last site-3 has active site volume of 87.46. Site-3 was too small to accommodate ligands, so they could not be potential pockets. Based on size of active site cavity site-1 & site-2 has been selected for further docking studies.

Table 1: Active sites of RBD

Virtual Screening (VS) with Lead Finder (LF)

After the identification of active sites, the VS was applied which allows a large chemical library to be screened computationally against a specific drug target. All the database compounds were subjected to docking using LF and the predicted binding poses were analysed. In the present study, we used Structure-based Virtual Screening (SbVS) which uses molecular docking techniques to screen large virtual libraries of database where all the molecules were docked in the RBD domain of spike protein. The compounds are scored based on the predicted interactions with the target protein and those with the top scores (hits) are selected for further analysis. All the database drugs were docked in the spike protein SARS-CoV-2 (PDB ID: 6M0J E- chain) by using Flare Docking module of Cresset software. For all the database molecules, less reasonable physicochemical parameters were discarded, leading to the selection of candidates with good drug-like properties. Later, a virtual library of approximately 3458 compounds extracted from Drugbank database were used for further screening using High-throughput virtual screening (HTVS) mode and top 100 molecules were finally docked in the "slow & accurate mode". All compounds were ranked based on their docking score values and those with a score < -8.0 and ligand having contact with more than one hydrogen bond in both the active sites were taken forward for final docking in "slow & accurate mode" of Cresset's Flare Docking software.

Molecular Docking and Interaction Analysis

Docking analysis and visualization of virtual screening hit sled to the identification of several drugs that bind selectively to the RBD site-1 and site-2. In total 17 drugs were identified based on rank score (RS), and binding energy (AG). These include some antivirals along with different set of approved drugs such as Triparanol, Homoharrigtone, Ritonavir, Astemizole, Amodiaquine, Fluspirilene, Pazopanib, Lapatinib, Levothyroxine, lodipamide, Cromoglicic acid, lohexol, Sitravinib, Dasatinib, Lapatinib Ertugliflozin which has shown most favourable binding affinity toward RBD S- protein. The docking results showed that some drugs bound to RBD site-1 and some bound to RBD site-2. The goal was to find those molecules which has great binding affinity towards both the active sites.

Detail binding analysis of selected drugs towards active site of spike protein SARS-CoV-2 was studied in detail. Interaction analysis of drugs with spike protein SARS-CoV-2 (RBD) were carried out to identify the compound having highest binding affinity with target protein.

Active site-1 composed of Arg454, Phe456, Arg457, Lys458, Glu465, Arg466, Asp467, Ile468, Ser469, Glu471, Thr473, Gln474 and Pro491 amino acid residues, while the site-2 composed of Leu335, Cys336, Pro337, Phe338, Gly339, Trp436, Phe342, Asn343, Val362, Ala363, Asp364, Val367, Leu368, Ser371, Ser373 and Phe374 amino acid residues, as shown in Figure 2. Active site-1 is more hydrophobic in nature as compare to active site-2.

The LF rank score is an indicator of the binding affinity of protein-ligand complex. The LF rank for selected drugs in the pockets 1 and 2 is described in Table 1 and 2 respectively. The binding orientation for each selected drugs having the least LF rank score, more negative LF rank score represent the better affinity of the drugs against target SARS-CoV-2 S-protein.

Among the docking studies performed on drugs, all the drugs had effective binding interactions with RBD domain of SARS-CoV-2 S-protein in both the active sites.

The LF rank score values were in the range of -8.88 to -12.5 in the active site-1, while in active site-2 the LF rank score range from -8.88 to -13.0 for all the identified drugs. The AG provide an accurate estimate of protein-ligand binding energy, on the assumption that the pose is correct. For the selected drugs the AG is range from -5.27 to -10.8 in site-1, while in site-2 the AG is range from -5.0 to -10.8. The number of hydrogen bond and the number of amino acid residues of SARS-CoV-2 (site-1 & site-2) interacting with each drugs are given in Table 1 and Table 2.

Table 2: Highest scoring molecules for SARS-CoV-2 Spike RBD in Site-1.

As shown in Figure 3, the graph represents the correlation between rank score and binding energies of all the selected drugs in RBD site-1 and site-2. All the drugs have more or less similar range of score and binding energies.

From the detailed docking analysis, it is observed that the drugs Lopinavir, Ritonavir, Homoharringtonine, Ertugliflozin, Sitravatinib and lodipamide have good LF rank score and AG in both the active sites. It is found that, these six drugs have formed strong H-bond contacts with more than two to three amino acid residues along with various hydrophobic interactions in spike protein resulting in increased binding affinity with target protein. Table 2 and Table 3 shows the rank score, binding energy, and list of interactions of the prioritized ligands with the site-1 and site-2 of the RBD.

Table 3: Highest scoring molecules for SARS-CoV-2 Spike RBD in Site-2

Detail binding analysis of Homoharrigtone revealed that it has hydrophobic interactions in the both the active sites. Docking of the Homoharrigtone within the active site of the RBD site-1 and site-2 is illustrated in Figure 4. In the site-1, it showed four hydrogen bonding with Arg457, Arg466, Asp467 and I Ie468 and well fitted in the hydrophobic pocket within the vicinity of

Phe456, Glu465, Ser469 and Glu471. Additionally, the cation-pi interaction between Lys458 constituted for a stable binding profile of the drug. In site-2, the Homoharrigtone is making hydrogen bonding with Ser371 and various non-bonded interactions. The drug is also involved in various hydrophobic interactions with Asn437, Thr438, Asn437, Thr438, Asn440, Asn343, Ala344, Thr345, Val367 and Leu368 amino acid residues.

Ritonavir drugs are anti retroviral medication for the treatment and prevention of HIV/AIDS. In the present study, this drug showed good binding with the RBD protein. The docking orientation of Ritonavir in site-1 and site-2 is representation in Figure 5. Ritonavir has good binding affinity towards RBD having a rank score of -10.3 in site-1 and -9.8 in site-2. The docking analysis of Ritonavir in the site-1 revealed that it has making three strong hydrogen bonding interactions with Arg457 and another with Asp467. The drug is also involved in one sulphuer bond with Asn460 and cation-pi interactions with Lys458. In the binding site-2, the drug was involved in hydrogen bonding interaction with Trp436 and Asn343. The compound was found placed proximal to various hydrophobic amino acids such as Leu335, Asn437, Thr438, Gly339, Asn440, Ile441, Phe342, Ala344, Leu368, Ser371 and Phe374 and hence exhibited hydrophobic interactions. Apart from these interactions the compound was further stabilized by pi-pi interaction with Phe338 and Phe373.

There is another drug Ertugliflozin which has most favourable binding affinity toward the RBD protein in both the sites. It belongs to the class of potent and selective inhibitors of the sodiumdependent glucose cotransporters (SGLT), more specifically the type-2 which is responsible for about 90% of the glucose reabsorption from glomerulus. The docking orientation of Ertugliflozin in site-1 and site-2 is represented in Figure 6. In site-1 Ertugliflozin drug in the RBD protein showed conventional hydrogen bonds with Asp467, 1 Ie468, Phe456 along with halogenic bond with Glu471 amino acid residue. The drug further stabilized by strong cation-pi interaction with Lys458. In the site-2, the drug was involved in hydrogen bonding interaction with Gly339, Asp364, Cys366, Arg509 amino acid residues. Additionally, it has pi-pi interactions with Phe374. The drug was found placed proximal to various hydrophobic amino acids such as Leu335, Trp436, Pro337, Phe338, Phe342, Val362, Ala363, Val367 and Leu368.

Another drug Sitravatinib which is under investigation in clinical trial NCT03680521 also showed good binding affinity towards RBD having a rank score of -12.5 in site-1 and -11.9 in site-2 (Figure 7). In site-1, Sitravatinib showed conventional hydrogen bodings with Cys480, Phe456, Arg457, Lys458, Asp467 and Gln474, amino acids, while in site-2 it was making strong hydrogen bonding interactions with Trp436, Asn343, Ser371. lodipamide, which is used as contrast agent for cholecystography and intravenous cholangiography is also showing good LF Rank score and AG in both the active sites. In site-1, the drug is making three hydrogen bonding interactions with Arg457, Lys458, and Arg466. Further, it is involved in strong cation-pi interaction with Arg457. Interaction analysis in active site-2 revealed that it has four hydrogen bonding interactions with Cys336, Phe342, Asn343, Trp436 amino acid residues. Further it makes a close contact with various hydrophobic interactions with Leu368, Asn440, Phe347, Arg509, Val367, Ser371 and Ser373 amino acid residues (Figure 8).

The RBD proteins were put through the functional assays as described in the foregoing Examples

Example 2: RBD-ACE2 biochemical assay

The above assay was performed using SARS-CoV-2 sVNT ready to use kit by Genscript, which is a competition ELISA, mirroring the viral neutralization process (Nature. Volume 583, pages 459 - 468, 2020). Drugs were used at a concentration of 50pM, lOOpM and 200pM. The reading was taken at 450 nm. The absorbance of the sample is inversely proportional to the inhibition of RBD's binding to ACE2 by the drug.

The data are shown as percent inhibition relative to binding observed in the absence of any drugs. As shown Figure 9, most drugs had some degree of inhibition thus corroborating the in silico predictions. Some drugs (Ritonavir, Dasatinib, Sitravinib, lodipamide, Ertugliflozin) showed dose response inhibition that approached more than 90% inhibition of binding of RBD to ACE2 and well correlating with in silico studies.

Example 3: Interleukin -1 B (IL-1 B) secretion assay

Human IL-1B ELISA was performed to detect the presence of IL-1B which is a key mediator of inflammatory response (Immunological reviews, 281(1), 8-27, 2018). The ELISA was performed using Human IL-1B high sensitivity kit from Invitrogen. The sample of ELISA used in the present study consist of cell culture media (referred to as sample hereafter) collected after treating the cells with drugs at lOOpM. Samples were added to the microplate precoated with human IL-1B antibody which captures the IL-1B present in the samples. A secondary anti-human IL-1B antibody conjugated to biotin was added to the plate. Following an overnight incubation, microplate was washed six times using wash buffer in order to remove any unbound biotin conjugated anti-human IL-1B antibody. Streptavidin-HRP was then added which binds to biotin conjugated antibody and the plate was incubated at room temperature on a shaker for an hour. After the incubation, plate was washed again following the same process as the previous wash step and an amplification reagent I was added to the wells. Following the incubation of 15 minutes and a wash, amplification reagent II was added. After incubation of half an hour in dark and a wash step later, a substrate solution was added turning the colour to blue. After 15-20 minutes, the reaction was terminated using a stop solution (turning the colour from blue to yellow). This final solution was read at 450nm. The OD of the sample is directly proportional to the amount of human IL-1B present in it.

To induce secretion of cytokine by the A549 lung cells, we used lipopolysaccharide/endotoxin (LPS) a known stimulus for inflammation seen in sepsis. We used LPS just to test the antiinflammatory properties of the drugs, since it is easily available, has robust and well established effects and does not need a BSL3 level biosafety level lab that would be needed for dealing with live virus. As shown in Figure 10, there was a modest but significant increase in IL-1 secretion by LPS relative untreated cells. When the drugs were added together with LPS some of them blunted the induced secretion whereas some drugs actually increased the secretion of this proinflammatory cytokine. Specifically, we would like to point out the inhibition by Ertugliflozin and lodipamide which have the potential to be repurposed for COVID-19. Applicants focused on Ertugliflozin since it is an approved drug for type-2 diabetes and has large body of safety data upon chronic use and has good potential for repurposing.

Example 4: Human Thrombodulin/BDCA-3 Immunoassay

Thrombomodulin (TBM, CD141 or BDCA3 fetomoduline) is a transmembrane glycoprotein expressed by a variety of cells, including endothelial cells (Journal of Thrombosis and Haemostasis, 2017, 25(5), 1020-1031) Human Thrombodulin ELISA was performed using a ready to use sandwich ELISA kit by R and D systems. Samples (media collected from our 3D vascular lung system) were added to a microplate precoated with monoclonal antibody specific for human thrombodulin. This was followed by an incubation of two hours (allowing any thrombodulin present in the sample to bind to the monoclonal antibody), the plate was washed with wash buffer four times. After washing an enzyme-linked monoclonal antibody specific for human thrombodulin was added to the plate. Another wash step was performed in order to remove any unbound antibody-enzyme reagent after completion of two hours of incubation. A colour substrate was then added turning the colour to blue. Reaction was quenched after 15-20 minutes using a stop solution which turned the colour from blue to yellow. Absorbance was read at 450 with wavelength correction set to 540 nm or 570 nm. Data is presented as amount of thrombomodulin in the media.

This protein was studied in detail in the context of more recent studies that showed uncontrolled blood clotting in COVID-19 patients which could contribute to mortality. LPS was used to induce the secretion of thrombomodulin in the 3D bioprinted vascular lung model. Almost all drugs had some degree of inhibition but Ertugliflozin in particular blocked the secretion of thrombomodulin by almost 40%, a very desirable property to possess for a drug that can be repurposed for COVID-19 (Figure 11).

Example 5: Vascular Lung Model

Cell Culture

For the 3D vascular lung model three types of cell were grown:

1. A549 cells were grown at 37° C in the growth medium DMEM (HIMEDIA #Cat No- AT007) supplemented with 10% (v/v) fetal bovine serum under the atmosphere containing 5% CO2. Cells were subcultured after reaching 80-90% confluence.

2. HUVEC cells were grown at 37° C in the Endothelial cell basal medium-2 (LONZA #Cat No-CC-3156 and CC-4176) under the atmosphere containing 5% CO ₂ . Cells were subcultured after reaching 80-90% confluence.

3. HL60 cells were grown at 37° C in the growth medium RPMI (HIMEDIA #Cat No- AT028) supplemented with 10% (v/v) fetal bovine serum under the atmosphere containing 5% CO ₂ . Cells were subcultured after reaching 80% - 90% confluence.

3D bioprinting In the 3D-vascular lung model four layers were bioprinted; first layer of collagen (30pl of Rat tail collagen) was added to each well in 96 well plate and incubated for 1 hour in CO2 incubator at 37°C. After incubation, once the collagen solidified, A549 cell flask with 80-90% confluence were trypsinized and cells were counted with the help of haemocytometer. Cell suspension was loaded into bioprinter syringe and 5xl0 ³ A549 cells were printed in each well of 96 well plate and the cells were incubated for 48 hours.

After 48 hours of incubation, previous A549 media was removed carefully and again 30pl of Rat tail collagen coating solution was added to each well in 96 well plate and incubated for 1 hour in CO ₂ incubator at 37°C. After incubation, once the collagen solidified, HUVEC cell flask with 80-90% confluence was trypsinized. Cell suspension was loaded into syringe and 5x103 HUVEC cells were printed in each well of 96 well plate with help of 3D bioprinter and the cells were incubated for 48 hours.

After 48 hours of incubation the media was removed and the final selected drugs from virtual screening were added with LPS at 0.5 pg/ml and without LPS at the desired concentration. Later, cells were incubated overnight. Next day the drugs were removed and the cells were washed once with media. Endothelial cells were fixed with 4% paraformaldehyde for 3 minutes at 40°C. After fixing with paraformaldehyde, the cells were again washed with media.

MTT stained HL60 monocytic cells lx 103 cells were added to each well in 96 well plate and incubated the plate for one hour and the wells were then washed twice with the media and pictures were taken of all the wells and the bound HL 60 cells were counted with the help of Image J, an open access software platform.

The effect of drugs on adhesion of HL60 monocytes to endothelium in our vascular lung model was examined. The recruitment of inflammatory cells such as monocyte is a common feature seen in COVID-19 as an early step of disease onset. This monocyte accumulation may greatly contribute to cytokine storm that is seen in this disease. As shown in Figure 12, none of the drugs were able to reverse the increase in monocyte adhesion stimulated by LPS. However, some drugs were able to reduce the increase caused by LPS. Again focusing on Ertugliflozin, it can be seen that it was able to reduce the impact of LPS. Form all the data it indicates that this drug (Ertugliflozin) has the potential to modulate monocyte accumulation observed during

COVID-19.

There is an urgent need to mitigate the early stages of COVID-19 caused by SARS-CoV-2 through drug repurposing. RBD of S-protein plays a pivotal role in viral-host attachment and one of the potential target for antiviral treatment against SARS-CoV-2. These specific structural domains can be targeted with small molecules or drug to disrupt the viral attachment to the host proteins. Repurposing approved pharmaceutical drugs provide an alternative approach for rapid identification of potential medication leads. In this study, FDA approved drugs were tested for their inhibitory properties towards the COVID-19 S protein (6M0J) using a virtual screening approach and computational chemistry methods. Rigorous in silico analysis through molecular docking analysis showed all 17 drugs have strong predicted binding affinities toward RBD domain in the SI protein. All these also showed good dock score and interaction with important amino acids along with various hydrophobic interaction and were well fitted in both the active sites of protein. More importantly the selection of drugs by computational modelling was corroborated in a series of in vitro studies. The most important confirmation came from the assay that examined the direct binding of RBD to ACE2. There was very significant correlation between the prediction from computational studies and the actual binding assay. We then conducted a series of studies that mimic some of the biological events seen in COVID- 19 such as secretion of I LI , presentation of a more thrombogenic endothelium by production of thrombomodulin and accumulation of inflammatory cells such as monocytes in the lungs. Applicants herein demonstrate that the drugs were able to favourably modulate these properties with some drugs being better than others. Ertugliflozin, a drug used for type-2 diabetes, in particular, possesses all desirable properties. This drug is an oral drug which works by blocking the activity of SGLT2 (Sodium-glucose co-transporter-2) that prevents glucose reabsorption in the kidney. This is chronic treatment drug which means that the safest of the drug has been thoroughly tested prior to approval. It is also readily available at an affordable price and could be repurposed immediately. To avoid any glucose lowering in COVID-19 patients by this drug, a nasal formulation can be developed which targets the lungs. This will also improve the potency and efficacy against the infection while requiring lower doses. While the development of vaccines will prophy lactica lly prevent the infection we do not have targeted drugs once the disease already sets in. Based on our work here we propose that ertugliflozin is a strong candidate for immediate repurposing for treatment of COVID-19 treatment. Applicants show that the disruption of the SARS CoV-2-RBD/ACE2 binding interface with high-affinity repurposed drugs as a promising strategy for preventing virus entry in human cells and paves the way for new COVID-19 treatment.