Human coronaviruses and therapeutic drug discovery

Background Coronaviruses (CoVs) are distributed worldwide and have various susceptible hosts; CoVs infecting humans are called human coronaviruses (HCoVs). Although HCoV-specific drugs are still lacking, many potent targets for drug discovery are being explored, and many vigorously designed clinical trials are being carried out in an orderly manner. The aim of this review was to gain a comprehensive understanding of the current status of drug development against HCoVs, particularly severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). Main text A scoping review was conducted by electronically searching research studies, reviews, and clinical trials in PubMed and the CNKI. Studies on HCoVs and therapeutic drug discovery published between January 2000 and October 2020 and in English or Chinese were included, and the information was summarized. Of the 3248 studies identified, 159 publication were finally included. Advances in drug development against HCoV, especially SARS-CoV-2, are summarized under three categories: antiviral drugs aimed at inhibiting the HCoV proliferation process, drugs acting on the host's immune system, and drugs derived from plants with potent activity. Furthermore, clinical trials of drugs targeting SARS-CoV-2 are summarized. Conclusions During the spread of COVID-19 outbreak, great efforts have been made in therapeutic drug discovery against the virus, although the pharmacological effects and adverse reactions of some drugs under study are still unclear. However, well-designed high-quality studies are needed to further study the effectiveness and safety of these potential drugs so as to provide valid recommendations for better control of the COVID-19 pandemic.

symptoms of COVID-19; patients infected with SARS-CoV-2 can develop dyspnoea a week after onset, and critical patients usually die from uncontrollable sepsis, respiratory failure, acute respiratory distress syndrome (ARDS) and septic shock [6]. Therapeutic interventions excluding virus-specific drugs, are often experiential or anecdotal and have not been tested in an integrated trial to provide sufficient and widely accepted evidence. The most common interventions include a combination of antivirals (such as ribavirin and lopinavir/ritonavir) and interferons (IFNs), corticosteroids, COVID-19 convalescent plasma and supportive treatment for critical patients [7].
Previous studies have revealed the invasion mechanism of HCoVs. In brief, S1 binds to the relevant receptor and induces endocytosis, then the conformation of the S2 subunit changes. The viral envelope fuses with the endosomal membrane and releases the nucleocapsid or viral genome [8]. Genomic RNA (gRNA) serves as a translation template for polyproteins pp1a and pp1ab, which are automatically hydrolysed into various non-structural proteins (NSPs), such as papain-like protease (PLpro), 3C-like protease (Mpro), and RNA-dependent RNA polymerase (RdRp). Full-length gRNA is replicated by negative sense intermediates and transcribed into subgenomic RNA (sgRNA). sgRNA encodes the structural proteins of the virus (N, M, E, and S) as well as helper proteins (e.g., 3, 4a, 4B, 5, and 8b). Particle assembly occurs in the ER-Golgi intermediate compartment (ERGIC) and is then released in the vesicle via the secretory pathway [3]. Interruption of the proliferation process might help cure patients and disrupt transmission. As the understanding of both the biological characteristics and pathogenicity of HCoVs has deepened, many potent targets for drug discovery have been explored, such as inhibiting HCoV invasion and strengthening host immune defences. In addition, traditional Chinese medicine might be effective in the fight against HCoVs. To establish additional evidence supporting recommended treatment strategies, some drugs, such as remdesivir, favipiravir, lopinavir/ ritonavir, arbidol/umifenovir, and hydroxychloroquine, have been tested in vigorously designed clinical trials. Herein, we review the progress in therapeutic drug discovery and development, including drugs that inhibit the CoV proliferation process (attachment and entry, replicase expression, replication, transcription and translation, assembly and release), antiviral drugs that affect the action of the host's immune system, and drugs derived from plants with potent activity, in order to accelerate drug discovery and development, especially during the current pandemic.

Search strategy
We searched two databases: PubMed (https ://www.ncbi. nlm.nih.gov/pubme d/) and CNKI (www.cnki.net). We searched for coronavirus (or HCoV) and important components (such as S protein, PLpro, Mpro, NSPs, RNA, N protein, E protein, host factors) and drugs (or Chinese medicine, plant derivates, or research or treatment). References of studies retrieved were cross checked as well. All the search results were evaluated. First, the titles and abstracts were screened to identify relevant studies; then, full texts were evaluated carefully to determine eligibility for inclusion. The complete search and selection processes were performed by two independent researchers. Any disagreements were resolved through consultation with a third researcher or team discussion until consensus was reached.

Inclusion criteria
(1) The target coronaviruses were HCoVs, with special attention to highly pathogenic HCoVs; (2) The studied drugs included newly developed targeted drugs, broadspectrum antiviral drugs, small-molecule compounds, plant derivatives, etc.; (3) the research performed included in vivo or in vitro tests, clinical trials, or literature reviews; (4) the publication language was English or Chinese; (5) the literature type was an article, review, or clinical trial; and (6) the publication time was from January 1, 2000, to October 27, 2020.

Exclusion criteria
(1) Duplicate studies; (2) studies for which the full text was unavailable; (3) news, reports, interviews, comments, patents, letters, or case reports; and (4) reviews or studies with the aim of elucidating the impact of coronavirus infection on the underlying diseases and their treatment in a target population.

Data extraction, summary, and analysis
We classified the selected documents according to the following categories: (1) antiviral drugs intended to inhibit the HCoV proliferation process; (2) antiviral drugs that affect the action of the host's immune system; and (3) antiviral drugs derived from plants with potent activity. All articles were processed using NoteExpress V 3.0 (Beijing Aegean Technology Co., Ltd., Beijing, China).

The scoping process
A total of 3248 records were retrieved. After excluding 322 duplicate records, 228 records with unavailable full texts, and 2539 records that met the exclusion criteria mentioned above, 159 records were finally included in this review. A flow diagram of the study selection process is shown in Fig. 1.
The proliferation process as well as the key targets of CoVs are presented (Fig. 2). A summary of the progress in therapeutic drug discovery and development targeting HCoVs is demonstrated below and is divided into three sections: drugs that inhibit HCoV proliferation (attachment and entry, replicase expression, replication, transcription and translation, assembly and release), antiviral drugs that affect the actions of the host's immune system, and drugs derived from plants with potent activity.

Blocking the HCoV proliferation process is the key to identify effective drugs against the virus
The drugs that target the CoV proliferation process are summarized in Table 1.

Inhibition of attachment and entry
The process of invasion can be divided into receptor binding and endocytosis [2]. Drugs developed from viruses or host structures that participate in the above two processes have the potential to block virus invasion.
Prevention of receptor binding The receptor binding domain (RBD) is the domain that binds to the receptor during HCoV invasion; RBDs have substantial diversity [2]. There have been many studies on therapeutic monoclonal antibodies (mAbs), including m336, m337 and m338, that target the RBD to prevent MERS-CoV invasion [9] and human-derived SARS-CoV [10,11]. In addition to mAbs, receptor-derived proteins based on the ligand-binding domain, such as P4 and P5 peptides, can be utilized to competitively bind to the RBD [12].
To enter a cell, the RBD needs to bind to a receptor. Thus, in theory, drugs that compete with RBDs for receptor binding sites, such as N-(2-aminoethyl)-1-aziridineethanamine (NAAE) [13] and anti-dipeptidyl peptidase 4 (DPP4) mAbs, can block CoV invasion [14,15]. However, considering that the receptors on the host cell surface also play an important role in the normal metabolism and function of the cell, the development of such drugs should take into account their impacts on the body, such as hypotensive and hypoglycaemic effects. Griffithsin (GRFT) can specifically bind to the glycosyl groups of protein S, thereby inhibiting virus invasion [16]. Prevention of endocytosis Inhibitors of host proteases that cleave protein S, such as transmembrane protease serine 2 (TMPRSS2) inhibitors (camostat [17], nafamostat [18]) and cathepsin L inhibitors (teicoplanin, dalbavancin), can prevent exposure and insertion of the hydrophobic end of S2 into the endosomal membrane [19]. Drugs targeting HR1/HR2, such as HR2P, P1 peptide, 229E-HR1P, 229E-HR2P and OC43-HR2P peptide [20][21][22][23], can prevent the formation of the 6-helix bundle structure, thereby inhibiting the fusion of the viral envelope with the endosomal membrane. Furthermore, arbidol [24], LJ001 [25], and NHR trimeric coiled coil alpha-helical lipopeptides prevent enveloped virus-cell membrane fusion [26]. Arbidol was tested in a clinical trial and appeared to reduce the SARS-CoV-2 RNA load [27]; however, other studies revealed that Arbidol did not improve the clinical outcomes of patients or SARS-CoV-2 elimination [28,29]. Attention should be paid to host factors that affect endocytosis. For instance, chloroquine inhibits endosomal acidification [30,31], and chlorpromazine inhibits clathrin-mediated endocytosis [30].
Recently, chloroquine phosphate has been recommended by Chinese scholars for the treatment of SARS-CoV-2 [32], but some studies have shown that hydroxychloroquine induces cardiotoxicity [33,34].

Inhibition of replicase expression
In the replication cycle, CoV RNA is first translated into two polyproteins, pp1a and pp1ab, which are then hydrolysed to generate sixteen NSPs with various functions [2]. Certain NSPs are essential for virus replication and transcription. Papain-like protease (namely, NSP3 and PLpro) and Achilles' heel 3C-like protease (also known as NSP5 and Mpro) play a vital role in hydrolysing polyproteins to generate NSPs [2]. Hence, inhibitors of the two proteases can block the generation of NSPs.

Inhibition of replication, transcription and translation
CoV replicase synthesizes the full-length antisense genome using gRNA as a template and then synthesizes new gRNA according to the sequence of the antisense RNA. Thereafter, with the help of RNA polymerase and certain transcription factors, the virus recognizes specific transcriptional regulatory sequences (TRSs) with "discontinuous transcription" and selectively transcribes all components that make up a mature mRNA. Finally, mRNA is translated into a variety of structural proteins (nucleocapsid protein N, membrane protein M, envelope protein E, and spike protein S) and accessory proteins (such as 3, 4a, 4b, 5 and 8b) [2].
Essential NSPs A variety of NSPs play significant roles in the replication process, whereas drugs targeting them are limited. NSP12 (RdRp inhibitors): The NSP12 (RdRp) inhibitor remdesivir prevents viral replication and thus reduces the viral load in patients [50]. However, two recent clinical trials have reached two different conclusions. A study published in the Lancet revealed that the drug is not effective [51][52][53], while a study published in New England Journal of Medicine showed that the drug shortened the length of hospitalization and virus removal time. However, this paper was withdrawn for many reasons [54]. Other drugs include a series of doubly flexible nucleoside analogues [55], galidesivir (BCX4430) [56], a novel synthetic adenosine analogue, 6′-fluorinatedaristeromycin analogues [57], and favipiravir. Favipiravir has been associated with improvement in chest CT findings [58,59]. However, the broad-spectrum antiviral drug ribavirin had no significant effects on clinical outcomes when administered alone for the treatment of SARS [60].
NSP16 [S-adenosyl-l-methionine (SAM)-dependent 2′O-MTase]: Drug action mechanisms can be divided into two types: direct termination of 2′O-MTase activity through the alteration of SAM (drugs that utilize this mechanism include S-adenosine-1 homocysteine, paclitaxel, and aurintricarboxylic acid (ATA) [66]) or alteration of 2′O-MTase activity by interfering with the binding of NSP16 to NSP10 (drugs that utilize this mechanism include complementary reverse peptides designed according to the sequence of the NSP16 binding domain [66]).
Viral nucleic acids and RNA synthesis complex Various small interfering RNAs (siRNAs) can interfere with viral replication as well as the expression of structural proteins and accessory proteins [71][72][73][74][75][76]. Mycophenolic acid may inhibit viral nucleic acid synthesis [77], but it is advisable to combine it with an interferon since its immunosuppressive effect may create an environment amenable to virus replication and dissemination. In addition, a synthetic chimaeric DNA-RNA hammerhead ribozyme can suppress the expression of SARS-CoV RNA [78]. Moreover, K22 can suppress RNA synthesis by inhibiting the formation of double membrane vesicles (DMVs) [79]. Finally, given the existence of replication intermediates, dsRNAactivated caspase oligomerizer (DRACO) can selectively induce apoptosis in cells containing viral dsRNA [80].
Protein N Newly generated RNA needs to bind to protein N to form a nucleocapsid for stability; protein N also plays an important role in the normal replication and transcription of gRNA [2]. Therefore, drugs targeting protein N, such as fibronectin-based intrabodies and the inhibitors PJ34 and resveratrol, may influence these processes [81][82][83].

Inhibition of assembly and release
Viral assembly occurs in the ERGIC, where proteins M and E play important roles [2]. Hexamethylene amiloride [84] blocks the E protein ion channel. CoV particles in ERGIC are transported through the secretory pathway in vesicles and released through exocytosis [2].
The interactions between the cytoskeleton and structural proteins are essential for the assembly and release of CoVs [3]. For example, nocodazole may reduce the amount of transmissible gastroenteritis virus (TGEV), which belongs to the genus α-CoV and shares a similar assembly and release mechanisms with HCoV-229E and HCoV-NL63, particles released from the body [85]. Nonetheless, the advantages and disadvantages must be considered before administering the drug due to the significant role of the cytoskeleton in the normal metabolism and functioning of cells.

Drugs that affect the action of the host's immune system could help relieve the symptoms Innate immunity
Complement activation and IFNs are believed to play an active role in the innate immune response against HCoVs.
Complement activation Inhibition of complement activation alleviates acute lung injury induced by SARS-CoV and MERS-CoV infection. For instance, anti-C5aR antibody treatment resulted in decreased viral replication in lung tissues in hDPP4-transgenic mice infected with MERS-CoV. SARS-CoV-infected C3 −/− mice exhibited significantly less weight loss and less respiratory dysfunction despite an equivalent viral load in the lungs [86][87][88].
IFNs. IFN-α/β (IFN-1) is an important component of innate immune defence, which protects mammalian hosts from viral infection [89]. While mild HCoV infections, such as infection by HCoV-229E, typically induce a high level of IFN-I production [90], SARS-CoV and MERS-CoV were shown to suppress the activation of the host innate immune response by inhibiting interferon production or signalling. Several structural proteins (M and N) [91][92][93], NSPs (NSP1 and NSP3) [94][95][96], and accessory proteins of SARS-CoV and/or MERS-CoV were identified as IFN antagonists [92]. In addition to inhibiting CoV replication, drugs targeting these proteins may work by unblocking IFN suppression by the CoV. IFN has been clinically indicated to be effective for the treatment of SARS-CoV and MERS-CoV. In clinical treatment, the routine use of IFNs is not recommended for SARS-CoV treatment [97]. IFNs are usually administered in combination with other drugs, such as IFN-β-1b combined with lopinavir/ritonavir [98] or ribavirin and IFN-α combined with lopinavir/ritonavir [99], for MERS-CoV treatment. In severe to critical COVID-19 patients, early treatment with IFN-α2b can reduce in-hospital mortality, but it has no significant benefit in moderately ill patients [100].

Cell-mediated immunity
Lymphocytopenia is commonly observed in patients infected with SARS-CoV [97], MERS-CoV [101], or SARS-CoV-2 [102], but the mechanism remains unclear. Human T cells are highly susceptible to MERS-CoV infection. Studies have demonstrated that MERS-CoV persists in T cell-deficient mice but is cleared in B celldeficient mice, suggesting that T cells play a critical role in MERS-CoV clearance [103]. SARS-CoV-specific T cells also play important roles in the recognition and clearance of infected cells [104].

Convalescent plasma
Convalescent plasma therapy may be beneficial for patients with early SARS infection because it provides antibodies from convalescent patients [118], but evidence of its efficacy in MERS-CoV patients is still lacking. It is recommended for the treatment of rapidly progressing, severe and critical cases of SARS-CoV-2 infection [99], but it is limited by safety concerns and inadequate sources. Trials indicate that convalescent plasma is most effective in reducing mortality when administered in the early stage of infection, but it does not significantly shorten the time to recovery [119,120].

Glucocorticoids
Corticosteroids not only suppress lung inflammation but also inhibit immune responses and pathogen clearance. Available observational data suggest impaired clearance of SARS-CoV and MERS-CoV as well as increased complication rates in survivors receiving corticosteroid therapy. Therefore, it is not advisable to administer corticosteroid treatment in patients with SARS-CoV-2-associated lung injury or shock outside of a clinical trial setting [121]. Recent clinical trials suggest that early, low-dose methylprednisolone administered in the short term improved clinical outcomes and reduced mortality in severe COVID-19 patients [122][123][124]. Guidelines from China recommend that glucocorticoids should be used in the short term as appropriate in patients with progressive deterioration of the oxygenation index, rapid radiographic development, and excessive activation of the inflammatory response [99].

IL-6 receptor inhibitors
IL-6 plays an important role in the development of a cytokine storm. As an IL-6 receptor inhibitor, tocilizumab does not prevent the disease from progressing, but it can reduce the symptoms of serious infection [125][126][127].
Clinical trials of drugs targeting SARS-CoV-2 are summarized in Table 2.
HCoV-NL63 Caffeic acid, which is related to the ethanol extract of Sambucus FormosanaNakai [156], has been confirmed to have a significant inhibitory effect on the invasion of HCoV-NL63, possibly by directly interfering with the binding of HCoV-NL63 to ACE2 and co-receptors, such as heparin sulphate proteoglycan.

SARS-CoV
The Ministry of Science and Technology of China has announced eight Chinese medicines that have been clinically confirmed to improve symptoms in SARS patients: Qingkailing injection, Houttuynia cordata injection, Radix isatidis granules, Xinxue granules, Jinlian Qingre granules, Dengzhanxixin injection, compound Kuh-seng injection and Xiangdan injection. In addition, Qingqi Liangying oral liquid and Qingwen oral liquid, and Jiedu pills, as well as anti-SARS I and anti-SARS II showed effective inhibitory effects on SARS-CoV [157] (Table 3).

Discussion
As of November 13, 2020, SARS-CoV-2 had infected 53 218 786 people worldwide and killed a total of 1 301 631 people. Unfortunately, the epidemic is still not under control in many countries. Despite a lack of HCoV-specific drugs, many potent targets for drug discovery have been explored, and many vigorously designed clinical trials are being carried out in an orderly manner. In the present study, we analysed the pathogenesis of and drug therapy targeting seven HCoVs, including four common types (HCoV-229E, -OC43, -NL63, -HKU1) and three highly pathogenic types (SARS-CoV, MERS-CoV, SARS-CoV-2); special attention was given to SARS-CoV-2.
Among the highly pathogenic CoVs, SARS-CoV transmission has been rare since 2004, so clinical trials of drugs and vaccines are difficult to carry out. To date, there are no specific drugs or vaccines against MERS-CoV. mABs, such as m336 [9], lopinavir/ritonavir [30], IFN [98,99], etc., are potential antiviral drugs against MERS-CoV, but additional evidence is needed to determine their efficacy.
Because COVID-19 is a new, acute, severe infectious disease, the anti-SARS-CoV-2 drug development strategies are to screen existing drugs to identify potentially effective drugs, to expand indications and to develop a vaccine. The safety of conventional drugs has been mostly verified; if effective, they can be quickly applied in clinical practice. To date, thousands of clinical trials of SARS-CoV-2 have been registered worldwide. Hot topics include antiviral drugs such as RaRp inhibitors [51][52][53][54]59], Mpro inhibitors [48,49], chloroquine and its derivatives [33,34], viral envelope inhibitors, arbidol [27,29], and immunotherapy drugs such as IFNs [100] and cytokine storm inhibitors [122,123,[125][126][127]. Usually, the duration from initial experimental research to clinical trial completion is long. However, due to the COVID-19 pandemic, many drugs have been entered into clinical trials that are not randomized, controlled, or double-blinded. Their efficacy, toxicity, and side effects are discovered during application. For example, it was previously reported that hydroxychloroquine and chloroquine acted against coronavirus, and the synergistic use of hydroxychloroquine and azithromycin reduced the viral load and improved clinical results. However, later studies found that the heart-related side effects of these drugs included extension of the QT interval, so the WHO terminated the studies [33,34]. Clinical trials found that lopinavir/ritonavir had a poor effect on COVID-19, while others, such as arbidol [27,29], remdesivir [51][52][53][54], favipiravir [59], IFN-α2b [100], convalescent plasma [119,120], corticosteroids [122,123] and tocilizumab [125][126][127], had different and even opposite results, which can be further validated by experimental evaluation and clinical experience. When the production of inflammatory factors is increased, convalescent plasma, corticosteroids, and tocilizumab should be used early and in appropriate amounts. Because most traditional Chinese medicines are compounds and few single drugs or single active ingredients are used, it is difficult to determine which ingredients are effective in clinical trials. It is hard to differentiate the compounds associated with the mechanism. In a laboratory study of a single active ingredient, glycyrrhizin had a strong inhibitory effect on SARS-CoV-2 [128,129], which is of great significance for further clinical study. The future research direction for traditional Chinese medicine is to identify and modify a single potent drug or active ingredient and adjust the compound dose and administration method.
This review summarized the conventional and potential drugs at according to each action site, which can improve clinicians' understanding of the results of current clinical studies to guide clinical decisions. It also enables researchers to understand drug action sites to discover potential effective drugs.

Houttuynia cordata
Polysaccharides and ethyl acetate extracts, such as afzerin and quercetin Act on the body's immune system with anti-complement activity, and inhibits 3CLpro and RdRp activity [133,134,136] Rheum rhabarbarum

Emodin
Suppress the interaction between S protein and ACE2 and inhibits 3clpro activity [135] Polygonum multiflorum N/A Suppress the interaction between S protein and ACE2 [136] Tripterygium regelii

Conclusions
This review summarized the progress in drugs that inhibit the HCoV proliferation, affect the action of the host's immune system as well as plant-derived Chinese medicines, which not only provides researchers a more comprehensive understanding of the current status of drug development against HCoVs, but also provides directions for further exploration. However, the pharmacological effects and adverse reactions of some drugs under study are still unclear, and hence well-designed high-quality studies are needed to further study the effectiveness and safety of these potential drugs in order to accelerate drug development targeting SARS-CoV-2 and thus promote progress towards ending the pandemic.