Introduction

A form of pulmonary disease was first reported in China from a city called Wuhan in the Hubei Province on December 31, 2019 [1]. The deadly disease was later termed as COVID-19 by the World Health Organization (WHO) on February 11, 2020. The identified causative novel coronavirus (2019-nCoV) is termed as severe acute respiratory syndrome-related coronavirus SARS-CoV-2 as it shares around 79.6% of genome similarity with SARS-CoV which also previously emerged in China during 2002–2003 [2]. With the announcement of COVID-19 as ‘Global Pandemic’ by WHO on March 11, 2020, SARS-CoV-2 has eventually affected 212 countries and territories around the world and 2 international conveyances. As of August 13, 2020, 20,881,635 cases have been confirmed with 748,503 deaths and 13,771,549 recovery cases, while among the active cases, 6,297,028 cases are in mild condition and 64,555 cases in a serious or critical condition [3] (Fig. 1a).

Fig. 1
figure 1

a The global trend of COVID-19 reported death, recovered and active cases till August 13, 2020; b structure of SARS-CoV-2

The literature reported seven coronaviruses (CoVs) that are known to cause human disease where the strains 229E (α-CoV), HKU1 (β-CoV), OC43 (β-CoV) and NL63 (α-CoV) caused mild infections of the upper respiratory tract in humans [4]. On the contrary, other two strains SARS-CoV (occurring in 2002–2003) and MERS-CoV (Middle East respiratory syndrome occurring in 2012) and the newly identified SARS-CoV-2 belonging to β-CoV have caused serious health threat and fatality [5]. The present scenario and available pathophysiology specify that SARS-CoV-2 is highly transmittable and contagious than its progenitor affecting not only the respiratory system but also the gastrointestinal system, central nervous system, kidney, heart and liver leading to multiple organ failure [6]. The SARS-CoV-2 spike S glycoprotein has 72% identical sequence with human SARS with a unique furin-like cleavage site, which is absent in other SARS-like CoVs [7]. The Cryo-EM structural evidence has revealed that SARS-CoV-2 has 10–20 times higher binding affinity to the ACE2 receptor than SARS-CoV which may lead to higher transmission and contagiousness [8]. Therefore, blocking of the isolated viral S protein at its host receptor region and/or binding within the S protein-ACE2 interface are the two most important strategies to design probable drugs for COVID-19 (Fig. 1b) [9]. The SARS-CoV-2 virus replicates via multiple processes after entering into the host cell, and the proteins associated with these replication steps are the principal targets to treat the infected patients by blocking the viral replication. The replication-associated proteins are [10]:

  1. a.

    Translation of genomic RNA,

  2. b.

    Proteolysis of the translated polyprotein with viral 3C-like proteinase,

  3. c.

    Replication of genomic RNA with the viral replication complex which comprises 3′-to-5′ exonuclease, RNA-dependent RNA polymerase (RdRp), endoRNAse and helicase, 2′-O-ribose methyltransferase,

  4. d.

    Assembly of viral components.

Along with the above-mentioned targets, the most commonly employed drug targets and drug discovery strategies employed all over the world right now are illustrated in Fig. 2.

Fig. 2
figure 2

a Possible drug targets and b drug development strategies to fight COVID-19

Presently, there is no specific treatment or approved drugs available to treat COVID-19. In most of the active cases, physicians are relying on symptom-based treatment for mild cases and primarily oxygen therapy (if required, with ventilator support) for critically ill patients. A set of approved marketed drugs like hydroxychloroquine (HCQ) [11, 12], chloroquine (CLQ) [12], combination of HCQ and azithromycin [13], remdesivir [14], lopinavir [15] and ritonavir [15] are being evaluated for the infection treatment; their clinical trials are also going on in different pharmaceutical industries. A series of new vaccines are also under clinical trial such as mRNA-1273 [16], Ad5-nCoV [17] and ChAdOx1 nCoV19 [18] along with existing Bacillus Calmette–Guerin (BCG) vaccine [19, 20] to check its efficiency in COVID-19. Due to severity and contagious nature of the SARS-CoV-2, researchers are exploring multiple in silico approaches and artificial intelligence [21,22,23,24,25,26] with the aim of identifying target-specific and potent therapeutic agents to speed up the discovery process. The RCSB protein data bank (PDB) (www.rcsb.org) has already deposited around 110 protein crystal structures associated with SARS-CoV-2 and COVID-19 to allow understanding important structural binding sites which can be explored in rational designing of small molecules.

The current review discusses the most updated and probable drug candidates which are being experimentally used to treat patients in different parts of the world. Also, their possible targets and pharmacological mechanisms of action which might not be clear in many cases and their pathophysiology along with the details about the status of convalescent plasma treatment and ongoing vaccine trials are discussed. We have compiled up-to-date in silico studies providing information related to computational tools, employed protein crystal structure used in the study followed by probable future drug candidates evolved from the repurposed virtual screening (VS) study employing docking, molecular dynamics (MD) and homology modeling. Therefore, the details related to SARS-CoV-2 transmission, protein structures, epidemiology, disease spectrum, diagnosis and testing are not discussed here at all as they have already been discussed in multiple literatures and separate reviews [1, 2, 4,5,6,7, 27,28,29]. The present review is significantly different from the other recently published ones on a similar topic in that it covers and gives emphasis on the in silico modeling studies in search of drugs against COVID-19. Thus, this paper provides an important source for the knowledge about possible drug candidates and vaccines along with their targets and pathophysiology information.

Investigational drugs or combination of drugs against COVID-19 up-to-date

At the time of writing this article, there are no clinically approved drugs or vaccine available for the treatment of COVID-19 [30]. However, there are many drugs under trial for the treatment of COVID-19 (chemical structures 119 in Figs. 3, 4, 5) including angiotensin II type I receptor (AT1R) blockers, antiviral drugs, antimalarial drugs, interferon, IL-6 inhibitors, corticosteroids, ascorbic acid, some antibacterial antibiotics, etc. An up-to-date list of drugs under trials against COVID-19 with their targets, mechanisms of action, the developer companies or institutions, uses and recent status are tabulated in Table 1. Several industries and research institute are also trying to develop miscellaneous drugs and/or therapeutics agents, followed by investigation of the effectiveness of combination of drugs listed under Box 1. Among these tabulated drugs, most effective ones are discussed here. Mechanisms of different categories of drugs used in COVID-19 patients on various stages of SARS-CoV-2 life cycle are schematically depicted in Fig. 6.

Fig. 3
figure 3

Structures of antivirals to combat COVID-19 (Compounds 18)

Fig. 4
figure 4

Structures of angiotensin II type I receptor (AT1R) inhibitors (Compounds 910), type 2 transmembrane serine protease (TMPRSS2) inhibitors (Compound 11) and antimalarials (Compounds 1213) to combat COVID-19

Fig. 5
figure 5

Structures of miscellaneous drugs to combat COVID-19 (Compounds 1419)

Table 1 Drug candidates under trial against COVID-19 with probable targets, mechanism of action, pathophysiology, application and current status [11, 12, 14, 30, 40, 43, 45, 53, 55, 57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88]
Box 1 Miscellaneous therapeutics under investigation for COVID-19
Fig. 6
figure 6

Mechanisms of action of different categories of drugs used in COVID-19 patients acting on various stages of the SARS-CoV-2 life cycle

The entry of SARS-CoV-2 to the host cell can occur in two ways, i.e., either via plasma membrane fusion or via endosomes (endocytosis blockers: CLQ and HCQ) (Fig. 6). In both ways, spike proteins (S1, S2) of SARS-CoV-2 mediate attachment to the membrane of a host cell and engage angiotensin-converting enzyme 2 (ACE2) as the entry receptor. Inhibitors like convalescent plasma, monoclonal antibodies bind to the spike glycoprotein, thus preventing the viral entry. When virions are taken up into endosomes, the spike protein can be activated by the cellular serine protease TMPRSS2 in close proximity to the ACE2 receptor, which initiates fusion of the viral membrane with the plasma membrane. Camostat mesylate inhibits the TMPRSS2 receptor. The plasma membrane fusion entry is less likely to trigger host cell antiviral immunity and therefore more efficient for viral replication. After the viral RNA is released into the host cell, polyproteins are translated. The coronavirus genomic RNA encodes non-structural proteins that have a critical role in the synthesis of viral RNA and structural proteins which are important for virion assembly. First, polyproteins are translated and cleaved by some of proteases like 3CLpro, PLpro, etc. (lopinavir, ritonavir and darunavir act as inhibitors of this step) to form RNA replicase-transcriptase complex. The non-structural protein RdRp is responsible for replication of structural protein RNA. (Remdesivir, favilavir and ribavirin act as inhibitors of this enzyme). Structural proteins S1, S2, envelope and membrane are translated by ribosomes that are bound to the endoplasmic reticulum (ER) and presented on its surface as preparation of virion assembly. The nucleocapsids (N) remain in cytoplasm and are assembled from genomic RNA. They fuse with the virion precursor which is then transported from the ER through the Golgi apparatus to the cell surface via small vesicles. The mature virions are then released from the infected cell through exocytosis, and then, they search another host cell (Fig. 6).

Angiotensin II type I receptor (AT1R) blockers

SARS-CoV-2 has transmembrane spikes (S protein). The spikes attached to the lipid membrane of the coronavirus recognize a host cell to attach and infect it with its viral RNA. The attachment of the coronavirus S protein to angiotensin converting enzyme 2 (ACE2) at its cellular binding site promotes the entry into human cells. The S protein contains two subunits such as an N-terminal S1 subunit which is responsible for receptor-virus binding and a C-terminal S2 subunit which is responsible for the fusion of virus with cell membrane [31, 32]. The S1 subunit has two domains such as a receptor-binding domain (RBD) and an N-terminal domain (NTD). At the time of infection, at first coronavirus binds to the human cell through interaction between the cell ACE2 receptor and S1-RBD of coronavirus. As a result, conformational changes in the S2 subunit are triggered followed by virus-cell fusion and entry into the target cell. ACE2 is a natural protein present in the lungs and the intestine (epithelial cells), and in the heart and the kidneys (endothelial cells). ACE2 regulates the blood pressure by converting angiotensin molecules. It was found that the coronavirus which caused the SARS outbreak in 2002 also binds to the same ACE2 molecule, but in case of SARS-CoV-2, the binding affinity is 10 to 20 times more on human cells than the spike from the SARS virus of 2002, which makes it a suitable target for COVID-19. Due to the high affinity of this virus to human cells, it is spread without any difficulty from one person to another than the earlier virus [33].

The entry of SARS-CoV-2 to the host cells via binding with ACE2 enzyme leads to ACE2 down-regulation. As a result, angiotensin II is produced excessively by the correlated enzyme ACE1, while a lower amount of ACE2 is not capable of transforming it to the vasodilator heptapeptide angiotensin 1–7. Thus, expression of higher ACE2 resulting from frequently medicating COVID-19 patients with AT1R blockers may resist them against acute lung injury. This can be described by following two complementary mechanisms: (1) blocking the excessive angiotensin II-mediated AT1R activation caused by the viral infection, (2) upregulation of ACE2 decreasing angiotensin II production by ACE and enhancing the production of the vasodilator angiotensin 1–7 [34]. The role of ACE2 to enter the coronavirus into the host cell and mechanism of action of ACE2 inhibitors to control the COVID-19 are depicted in Fig. 7.

Fig. 7
figure 7

The role of ACE2 receptor for the entry of coronavirus into the host cell, and mechanism of action of ACE2 inhibitors and TMPRSS2 inhibitors to control COVID-19

The SARS-CoV-2 uses the S protein to facilitate viral entry into the host cells. The pathogen S protein consists of two subunits S1 and S2, of which “S1” allows entry of pathogen and binding of S protein to ACE2 (cellular receptor) (Fig. 7). In addition, the entry requires S protein priming by cellular proteases, which is responsible for cleavage of S protein. After this, “S2” subunit employs fusion of viral and cellular membranes. SARS-CoV-2 engages ACE2 as the entry receptor that can be blocked by ACE2 inhibitors and Arbidol, and it employs the cellular serine protease TMPRSS2 for S protein priming which is inhibited by camostat mesylate (Fig. 7). Conclusively, SARS-CoV-2/ACE2 interface occurs in the molecular level, and the efficiency of ACE2 usage is a key determinant of SARS-CoV-2 transmissibility.

Although it is widely accepted that coronavirus enters into the host cell through the ACE2 receptor, due to limited number of studies, it is yet to establish how ACE2, AT1 and AT2 receptors exert their activities in coronavirus-induced diseases [35, 36]. Thus, ACE2 inhibitors and AT1 receptor antagonists [e.g., L-163491 as a partial antagonist of AT1 receptor and partial agonists of AT2 receptor; losartan, valsartan, irbesartan, candesartan cilexetil, telmisartan, and eprosartan (FDA-approved AT1 receptor blockers)] may be used as important drug candidates to control lung injury of COVID-19 patients [34]. The binding of viral S protein with its receptor ACE2 on host cells followed by viral endocytosis into the cells may also be a possible drug target. For example, the broad-spectrum antiviral drug Arbidol recently entered the clinical trial for the treatment of SARS-CoV-2 which may act by inhibiting virus-host cell fusion, thus preventing the viral entry into host cells against influenza virus [37,38,39].

Camostat mesylate

The serine protease TMPRSS2 produced by the host cells plays a key role for cell entry of coronaviruses by S protein priming to the receptor ACE2 binding in human cells (Fig. 7). A recent study shows that camostat mesylate, a clinically approved inhibitor of TMPRSS2 (responsible for S protein priming), has been able to block SARS-CoV-2 infection of lung cells. Thus, this drug may be a potential drug candidate for COVID-19 [40].

Remdesivir

Remdesivir, a nucleoside analog and a monophosphoramidate prodrug of remdesivir-triphosphate (RDV-TP) developed by Gilead Sciences Inc. (USA), was previously tried for the Ebola virus disease, and it showed promising effects in MERS and SARS. It acts by inhibiting RNA-dependent RNA polymerases (RdRp). Incorporation of this drug into nascent viral RNA chain causes premature termination. For incorporation of remdesivir-TP into nascent viral RNA chains, it competes with adenosine-triphosphate. After incorporation into the viral RNA at position i, remdesivir-triphosphate arrests RNA synthesis at position i + 3. Due to the incorporation of 3 additional nucleotides after RDV-TP, it does not cause instant chain termination because these three additional nucleotides may protect the inhibitor from excision by the viral 3′–5′ exo-ribonuclease activity (Fig. 6). Recent reports showed that the EC90 value of remdesivir against COVID-19 in VeroE6 cells was 1.76 µM, half-cytotoxic concentration (CC50) was greater than 100 µM, and the selective index (SI) was greater than 129.87, suggesting that its working concentrations are likely to be achieved in nonhuman primate (NHP) [14, 41]. This drug is also able to inhibit virus infection proficiently in human liver cancer Huh-7 cells sensitive to COVID-19. A recent case study revealed that treatment with remdesivir improved the clinical condition of the first patient infected by SARS-CoV-2 in the USA [42]. A recent in vitro data showed that remdesivir and chloroquine (CQ) phosphate are capable of inhibiting SARS-CoV-2 infection [14]. Remdesivir is currently being studied in Phase-III clinical trials against SARS-CoV-2 in Wuhan, China, as on February 4, 2020, and in the USA.

Chloroquine/hydroxychloroquine

Chloroquine (CLQ) and hydroxychloroquine (HCQ) have received deep attention because of positive results from some small studies. An antimalarial drug, CLQ, has recently been reported to have potential in vitro activity against SARS-CoV-2. CLQ protects from viral infection by enhancing endosomal pH (making the environment unfavorable) which is required for virus-cell fusion. This drug may also block viral infection by inhibiting viral enzymes or processes like viral DNA and RNA polymerases, virus assembly, new virus particle transport, immunomodulation of cytokine release and virus release. CLQ also affects the glycosylation process of ACE2 (as discussed earlier that to enter the host cell, the viral S protein binds with this receptor) [11, 14, 43, 44]. Besides this mechanism, a recent report [12] showed that this drug also acts by inhibiting the sialic acid containing glycoprotein and gangliosides (act as primary attachment factors along the respiratory tract) mediated attachment to the S protein which is the first step for viral replication. In the NTD of the S protein of SARS-CoV-2, a ganglioside-binding site was recognized. The antimalarial drug CLQ was found to be a probable blocker of the S–ganglioside interaction. Thus, this drug may be used to fight pathogenic coronaviruses especially SARS-CoV-2 which is responsible for COVID-19. A detailed mechanism of action of CLQ and HCQ against SARS-CoV-2 is illustrated in Fig. 8. A recent report showed that the EC90 value of CLQ against the SARS-CoV-2 in VeroE6 cells was 6.9 µM, which may be clinically achievable. Although specific data are not available, this drug is able to inhibit the exacerbation of pneumonia patients with SARS-CoV-2 infection.

Fig. 8
figure 8

Schematic representation of different mechanistic pathways of CLQ and HCQ against SARS-CoV-2

Figure 8 explains two possible mechanisms of CLQ and HCQ against SARS-CoV-2. The mechanism 1 is that CLQ and its derivative HCQ are weak bases, which can raise the pH of acidic intracellular organelles, such as endosomes/lysosomes, essential for membrane fusion. On the other hand, mechanism-2 explains the entry of SARS-CoV-2 into the host cells also depends upon sialic acid (Neu5Ac) containing glycoproteins and gangliosides that act as the key binding factors along the respiratory tract. A ganglioside-binding site at the N-terminal domain (NTD) of the S glycoprotein of SARS-CoV-2 was recognized, and CLQ was found to be a possible blocker of the S–ganglioside interaction which occurs in the first step of the viral replication cycle (i.e., attachment to the surface of respiratory cells, intermediated by the S protein) [12]. The interaction was augmented by placing the negative charge of the carboxylate anion of Neu5Ac and one of the two positive charges of CLQ. SARS-CoV-2 especially interacted with 9-O-acetyl-N-acetylneuraminic acid (9-O-SIA). In this case, the carboxylic acid group of the sialic acid interacted with the cationic group of the nitrogen-containing ring of CLQ. The formed complex of CLQ and 9-O-SIA was further stabilized by OH-π and van der Waals interactions. Next, the complex developed from HCQ was very close to that obtained from CLQ, although numerous conformational adjustments happened for the period of the simulations. Interestingly, the –OH group of HCQ reinforced the binding of CLQ to Neu5Ac via formation of a hydrogen bond. The formed complex of CLQ-OH and 9-O-SIA will be stabilized again like CLQ to form a protective layer against fusion of the SARS–CoV-2.

HCQ is a hydroxy derivative of CLQ, which can block the viral infection by a similar mechanism as chloroquine; thus, this drug may also be a potential candidate against SARS-CoV-2. This drug is less toxic (~ 40%) than chloroquine in animals. It was found that there were seven clinical trials registered as on February 23, 2020, in the Chinese Clinical Trial Registry (http://www.chictr.org.cn), for using HCQ to treat COVID-19. The in vitro results suggested that this drug can efficiently inhibit SARS-CoV-2 infection. Based on an in vitro report, it was suggested that HCQ may be more potent than CLQ to treat COVID-19 [11, 44]. (EC50 values for CLQ > 100 µM at 24 h and 18.01 µM at 48 h; EC50 values for HCQ were 6.25 µM at 24 h and 5.85 µM at 48 h). Due to the relatively low selectivity index (SI) of HCQ, it requires cautious designing and conducting of clinical trials to attain resourceful and safe control of the SARS-CoV-2 infection.

Favipiravir or Favilavir (Avigan)

A purine nucleoside presently labeled as Avigan, developed by Fujifilm Toyama Chemical of Japan, has recently been approved for Phase-III clinical trial (March 31, 2020) for the COVID-19 patients. This drug is approved for manufacturing and sale in Japan for the treatment of influenza as an antiviral. In case of Influenza virus, it selectively inhibits RNA polymerase which is essential for viral replication when human cells are infected. It is believed that this drug may be effective for the treatment of COVID-19 as SARS-CoV-2 uses same enzyme (RNA polymerase) for replication and classification into the same type of single-stranded RNA virus like influenza. Thus, this drug acts by inhibiting the RdRp leading to inaccurate viral RNA synthesis (Fig. 6). This drug is recommended by the director of the China National Center for Biotechnology Development under the Ministry of Science and Technology to treat COVID-19. Italy has also approved the drug to treat COVID-19 cases. Due to the effectiveness of this drug against COVID-19, it is being mass-produced as generic version in China [45, 46].

Ritonavir and lopinavir (Kaletra)

These drugs are approved HIV-1 protease inhibitors, used in combination with other anti-retroviral drugs to treat HIV-1 infection in both adults and pediatric patients who is older than 14 days. Coronaviruses encode either two or three protease enzymes like papain-like proteases (PLpro), a serine-type protease, the main protease, or Mpro which cleave the polyproteins into non-structural polyproteins (nsps). These nsps are essential for viral RNA synthesis. Ritonavir and lopinavir act by inhibiting these protease enzymes. The mechanisms of action of these drugs suppressing coronavirus activity [47] are depicted in Fig. 6. A combination of these drugs is recommended in Italy to treat COVID-19 patients. Several trials are going on worldwide using these drugs or in combination of other drugs. A collaborative research from China and the UK conducted a clinical trial to examine the effectiveness of a combination of these two drugs against COVID-19 which was published in the New England Journal of Medicine (NEJM) [15]. The output of their trial did not provide any significant benefit in the patients with COVID-19. They suggest that “future trials in patients with severe illness may help to confirm or exclude the possibility of a treatment benefit.” Among those trials, two trials are investigating against pneumonia caused by COVID-19. One trial is conducted in the Tongji Hospital, Wuhan, China, using lopinavir–ritonavir against Arbidol hydrochloride (influenza drugs) and oseltamivir (NCT04255017). In South Korea, a comparative study of lopinavir–ritonavir against HCQ in patients with mild cases of COVID-19 (NCT04307693) was made. The two arms of the WHO SOLIDARITY trial are lopinavir–ritonavir alone and in combination with interferon-beta [48].

Ivermectin

Ivermectin is an FDA-approved broad-spectrum antiparasitic drug, which recently showed in vitro antiviral activity against SARS-CoV-2. It acts by inhibiting the interaction between HIV-1 integrase protein (IN) and the importin (IMP) α/β1 heterodimer which is responsible for IN nuclear import [49, 50]. Therefore, (IMP) α/β1 is unable to bind to the viral protein and preventing it from entering the nucleus, thus inhibiting HIV-1 replication [49, 50]. As a result, inhibition of the antiviral responses is reduced leading to a normal, more efficient antiviral response.

Monoclonal antibodies

The trial of potential monoclonal antibody-based therapy against COVID-19 is going on by using the previous knowledge on the neutralizing monoclonal antibodies (nMAb) against similar coronaviruses such as SARS-CoV and MERS-CoV. Monoclonal antibodies targeting the vulnerable sites of trimeric spike (S) glycoproteins on the viral surface which are responsible for the entry to the host cell are increasingly being recognized as a promising class of drugs against COVID-19. Potential neutralizing monoclonal antibodies mainly targeting the receptor-interaction domain at S1 subunit ultimately disabled cell–receptor interactions [51,52,53]. The detailed mechanism of action of this class of drugs is depicted in Fig. 9. Recently, several monoclonal antibodies, namely tocilizumab (atlizumab), mavrilimumab, lenzilumab, leronlimab, gimsilumab, sarilumab, siltuximab (Sylvant), camrelizumab (AiRuiKa), eculizumab (Soliris), etc., are being tried to investigate their potency against COVID-19 disease, and these are tabulated in Table 1.

Fig. 9
figure 9

Schematic diagram of the role of human monoclonal antibodies to block SARS-CoV-2

The human monoclonal antibodies (hmAbs) are developed by using several strategies against SARS-CoV-2 including preparation of hybridomas (using a transgenic mouse), phage display technologies and the immortalization of convalescent B cells. At present, the hmAbs production has been carried out with two strains of transgenic mice, i.e., Medarex HuMab-Mouse and Xeno-mouse [54]. The difference between these two strains of mice is that mouse L chain genes are still functional in the Medarex HuMAb-Mouse; thus, these mice can produce chimeric mAbs. The Xeno-Mouse from Amgen has all the mouse L chain genes deleted, and B cells produce only human Abs. One of these mice monoclonal antibodies [chimeric (Medarex HuMAb) (or) hybridomas (Xeno-mouse)] was immunized to produce neutralizing antibodies and further evaluated. The developed neutralizing antibodies bind to a specific portion to the RBD (N-terminal (amino acids 12–261, 130–150), and C-terminal of the RBD (amino acids 548–567, 607–627)) to prevent fusion of SARS-CoV-2 with the target cells.

COVID-19 convalescent plasma

Due to emergency of COVID-19 patients, recently, FDA has recommended to healthcare providers and investigators the use of convalescent plasma collected from individuals who have recovered from COVID-19 that may contain antibodies to SARS-CoV-2. However, till now, this therapy has not yet been shown to be effective and safe against this disease. Thus, it is very essential to investigate the safety and effectiveness of COVID-19 convalescent plasma in clinical trials. For this therapy, FDA has recommended some guidelines as follows [55]: (1) the pathways for use of investigational COVID-19 convalescent plasma; (2) patient eligibility; (3) collection of COVID-19 convalescent plasma, including donor eligibility and donor qualifications; (4) labeling, and (5) record keeping.

The pharmacological safety data including dose, drug–drug interactions and toxicities of selected potential drug candidates are provided in Table 2.

Table 2 Pharmacological safety data of selected potential drug candidates [11, 12, 14, 34, 38, 39, 43,44,45, 57,58,59, 64, 69, 70, 89]

Vaccines

Due to the worldwide outbreak of the COVID-19, the general public are keenly watching the progress of development of COVID-19 vaccines [56]. Dr. Anthony S. Fauci, Director of National Institute of Allergy and Infectious Diseases (NIAID), said that “finding a safe and effective vaccine to prevent infection with SARS-CoV-2 is an urgent public health priority.” But development of a new vaccine against a new disease is not an easy task. Various research institutes and industries are giving their full efforts to develop a vaccine against this pandemic disease with its earliest. Thankfully, the progress is rapid due to various reasons such as: (1) sharing the efforts to sequence the genetic material of SARS-CoV-2 by China throughout the world, (2) coronaviruses were already on the radar of health science researchers, (3) the knowledge from SARS and MERS caused by corona viruses and (4) also learnings from the vaccines against SARS and MERS which were stopped or postponed when those outbreaks were controlled may still be used to defeat COVID-19. The progress of most promising vaccines against the pandemic disease COVID-19 being made by various industries and research institutes is tabulated in Table 3. Several miscellaneous vaccines under investigation for COVID-19 are listed under Box 2.

Table 3 The progress in development of vaccines by different companies and institutes throughout the world [16,17,18,19,20, 90, 91]
Box 2 Miscellaneous vaccines under investigation for COVID-19 [92, 93]

In silico modeling applied to search the future drug candidates

To combat the COVID-19 pandemic, researchers must fight with the time to save as many lives as possible. To save the time and speed up the drug discovery process, in silico modeling provides one the best possible options. Till now, in most of the cases, researchers are trying the computational repurposing theory of existing approved drugs (synthetic as well as from natural origin) for SARS-CoV-2 employing docking, homology modeling and molecular dynamics (MD) studies to identify probable magic drugs for COVID-19. Huang et al. [21] computationally designed a short protein fragment or peptide which may block the coronaviruses’ ability to enter human cells by binding to the viral protein which is one of the first kinds of peptide treatments routing for experimental efforts.

Smith and Smith [22] analyzed 8000 small drug molecules and natural products (SWEETLEAD library database) employing restrained temperature replica-exchange MD simulations combining virtual screening through the ensemble docking to identify the effective drug for COVID-19 which might stop the virus by two ways: (a) disrupting S protein and ACE2 receptor interface stability; or (b) by troubling the capability of the S protein to recognize the ACE2 receptor. The simulation was performed in the IBM’s supercomputer SUMMIT which is recognized as the world’s most powerful computer with 200 petaflops speed. The authors reported the predicted binding affinities of all studied molecules for isolated SARS-CoV-2 S protein and the S protein human ACE2 receptor interface, followed by proposed 77 small molecules by docking studies (47 ligands employing interface docking and 30 ligands identified by the isolated S protein docking). Among the identified 77 molecules, the authors finally reported top 7 regulatory approved small molecules [Pemirolast (Zinc ID: 5783214), Isoniazid pyruvate (Zinc ID: 4974291), Eriodictyol (Zinc ID: 58117) and Nitrofurantoin (Zinc ID: 3875368) binding within the S protein-ACE2 interface; Ergoloid (Zinc ID: 3995616), Cepharanthine (Zinc ID: 30726863), and Hypericin (Zinc ID: 3780340) binding with the S protein receptor] which may serve as possible drug candidates for COVID-19 (Fig. 10a).

Fig. 10
figure 10

Schematic workflows performed by Smith and Smith [22] (a) and by Ton et al. [23] (b)

Ton et al. [23] identified 1000 noncovalent inhibitors for SARS-CoV-2 main protease (Mpro) using 1.36 billion compounds from the ZINC15 library employing deep docking platform using Glide SP module which utilizes QSAR models trained on docking scores. The 4MDS protein with 1.6 Å resolution bound to a noncovalent inhibitor was used for the docking study, and among the identified 1000 noncovalent inhibitors, ZINC000541677852 is the top hit drug candidate (Fig. 10b).

α-Ketoamide inhibitors are proposed as new drug candidates by designing, synthesis, followed by a docking study on the main protease (Mpro, 3CLpro) which has a crucial role in processing the polyproteins that are translated from the viral RNA [24]. Three different PDB structures were employed for the study; they are 6Y2E, 6Y2F and 6Y2G. The authors modified previously designed inhibitor (1) (earlier used for beta-, alpha coronaviruses and 3C proteases of enteroviruses) by incorporating the P3-P2 amide bond into a pyridone ring to enhance the half-life of the newly designed (2) compound in plasma followed by replacing the hydrophobic cinnamoyl moiety with less hydrophobic Boc group (Fig. 11a). Although the inhibitory concentration decreased from 0.18 ± 0.02 µM (1) to 2.39 ± 0.63 µM (2), but molecule 2 reported three times higher plasma binding and 19 times better plasma solubility compared to molecule 1. Further, to improve the antiviral activity against betacoronaviruses of clade b, the authors replaced the P2 cyclohexyl moiety of molecule 1 by the smaller cyclopropyl fragment to produce compound 3 which showed IC50 value of 0.67 ± 0.18 μM for the purified recombinant SARS-CoV-2 Mpro.

Fig. 11
figure 11

Schematic workflows performed by Zhang et al. [24] (a) and by Zhou et al. [25] (b)

Zhou et al. [25] reported an integrative antiviral drug repurposing analysis employing pharmacology-based network medicine platform by a two-step process: (a) quantifying the relationship between the human coronaviruses (HCoV) and host interactome and (b) identifying the drug targets in the human protein–protein interaction network. As per phylogenetic analyses, SARS-CoV-2 has the highest nucleotide sequence identity (79.7%), followed by the envelope and nucleocapsid protein sequence identities of 96% and 89.6%, respectively, with SARS-CoV. Employing the network proximity analyses between drug targets and HCoV-associated proteins followed by gene set enrichment analysis (GSEA), it was possible to identify 16 probable anti-HCoV drugs (e.g., melatonin, mercaptopurine, and sirolimus, etc.). Later, the drugs were validated through HCoV-induced transcriptomics data in human cell lines and enrichment analyses of drug-gene signatures. The authors also reported three effective drug combinations among the probable hits identified through ‘complimentary exposure’ pattern (Fig. 11b).

Grifoni et al. [26] employed the Immune Epitope Database and Analysis Resource (IEDB) to characterize the sequence similarity between SARS-CoV and SARS-CoV-2 through homology modeling. The epitope prediction identified a priori potential B and T cell epitopes for SARS-CoV-2 which are the promising targets for immune recognition, followed by the discovery of diagnostics and future vaccines.

Within a short period of time, a huge number of in silico results were deposited in the preprint servers from all over the world and at the same time a few are already published. As in most of the cases, multiple in silico drug designing and virtual screening tools were employed for repurposing theory of approved existing drugs [94], thus, major information is gathered in Table 4 to avoid similar discussion several times [95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119].

Table 4 A comprehensive list of in silico studies in search of drug molecule to fight COVID-19 [95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119]

An interesting aspect is that most of the studies are based on repurposing theory of existing approved drug employing docking and MD supported VS. Majorly the authors relied on approved antiviral drugs along with DrugBank database, Zinc database, Natural compounds’ databases along with one of the most discussed molecules of recent time, HCQ. Thus, the basic ideas of implication of in silico tools and repurposing of approved drugs are same, but the only difference is selection of the target protein in different studies. Without any doubt, the crystal structure of COVID-19 main protease (Mpro) in complex with an inhibitor N3 (PDB: 6LU7) is the most accessed protein for the drug discovery. But based on the target type, the choice of proteins can be different. Thus, to assist researchers, we have classified the target proteins into 14 types covering 110 PDB crystal ID from PDB (https://www.rcsb.org/) available until April 12, 2020, and enlisted in Table 5.

Table 5 Type of target protein to identify the efficient drug molecule of SARS-CoV-2

Is interspecies modeling the future of COVID-19 drug discovery? An unexplored concept under in silico tools

Multiple incidents illustrated the outbreak of corona virus in animals, especially in pets [120, 121]. Animals from dogs, cats, tigers, lions to minks are already tested positive and showed mild to severe symptoms of corona virus, followed by death of few instances. Although these events are scattered and not enough to study in the middle of human crisis, we cannot ignore the fact of human to animal transmission. If human to animal transmission is true, then there is a possibility of mutations, insertions and deletions in the genome sequence of this deadly virus in these animals in future with probabilities of zoonotic transfer of a stronger form of present SARS-CoV-2 virus from them to human in the near future. Thus, these small incidents need to be checked very carefully to avoid any future transmission.

The above-mentioned facts help to build an interspecies analysis by Kar and Leszczynski [122] for animals to human transmission or vice versa to aid in the drug discovery process of COVID-19 in upcoming days. The in silico interspecies-quantitative structure–activity relationship (i-QSAR) modeling correlates and then extrapolates the response (activity/toxicity) data from animal source to human which will be helpful for drug discovery. Due to genome sequence similarity of SARS-CoV-2 with the pangolins and bat, experimental data of drug candidates to pangolins/bat along with structural and physicochemical properties of drugs can be correlated with human response endpoint which is the first step of modeling. In the next step, extrapolation of animal data to human data can lead to no human testing through developed i-QSAR model. Once the acceptable predictive i-QSAR model is ready, there might be no need of future animal testing. As bat and pangolins are endangered species, thus future introspection can be performed in dogs and cats due to their better accessibility as well as considering them possible carriers of SARS-CoV-2 from the recent incidents.

Conclusion

Researchers from all over the world are trying to find the medicines which will help to stop the transmission of the virulent SARS-CoV-2 virus, mitigate the symptoms of the infected patients and help to lower the death toll throughout the world. Unfortunately, till now there is no silver gunshot which can solve this pandemic COVID-19. We have reported here a comprehensive and updated discussion on drug or drug candidates under investigation with their probable targets and mechanisms of action which might not be clear for many cases earlier, along with the effort of ongoing vaccine trials, monoclonal antibodies therapy and convalescent plasma treatment. We have presented here the ongoing computational efforts related to in silico tools to explore the probable drug candidates against COVID-19 along with the up-to-date target protein information. The information provided from in silico approaches may work as a magic bullet for the medicinal chemists to accelerate the drug discovery and development process. Overall, this review provides a strong intellectual foundation to support progress of ongoing research related to thorough knowledge of drugs or investigational drugs, vaccines and in silico approaches which can be helpful for the development of new drug candidates for the treatment of COVID-19.