Design, synthesis and biological evaluation of 2-aminoquinazolin-4(3H)-one derivatives as potential SARS-CoV-2 and MERS-CoV treatments

Graphical abstract

Coronavirus is a single positive-stranded RNA virus that was discovered in 1960 while looking for a new cold virus that infects the upper respiratory tract. [1][2] The clinical significance was relatively low because of its weak symptoms. 3 However, with the outbreak of SARS-CoV in 2003, its clinical significance has received new attention. [3][4] SARS-CoV spread to 4 countries, with 8,422 confirmed cases and 916 deaths, with its mortality rate of 11%, raising the public health awarness. [4][5] In 2012, a new outbreak of MERS-CoV occurred and spread to 27 countries by January 2020, resulting in 2519 confirmed cases and 866 deaths, with a mortality rate of 34%. 6 Most recently, COVID-19, which was caused by SARS-CoV-2 outbreak in Wuhan, China, in December 2019, has spread worldwide, causing 40 million confirmed cases and 1 million deaths. 7 Eventually, on March 11, 2020, COVID-19 was declared as third pandemic after the 1968 Hong Kong flu and 2009 influenza by the WHO. [8][9] For treatments of MERS-CoV, it is generally recommended to use drugs such as interferon, immunomodulatory factor, and antiviral drugs such as ribavirin or lopinavir. 10 However, there are reports that interferon and ribavirin may cause side effects such as poor bone marrow function, anemia, and virus mutations. 11 In addition, although the monkey model showed a therapeutic effect 12 , it did not show a great effect in actual clinical trials, requiring the development of a safer and more efficient MERS treatment. 13 In the case of COVID-19, Remdesivir, which received urgent approval, as well as Nafamostat, and Hydroxychloroquine, are considered as promising therapeutic candidates. 14 However, due to side effects and low clinical effects, 15 safer and more effective treatments need to be developed.
The anti-SARS-CoV-2 and anti-MERS-CoV activities of the synthesized compounds were evaluated by immunofluorescent assay in Vero cells. 16,20 In this study, Vero cells were stained using antibodies targeting spike protein for MERS-CoV and nucleocapsid protein for SARS-CoV-2, and the infection rate was measured by imaging the infected Vero cells through microscope.
In the next phase of optimization, we evaluated substituent effects of 5 to 8 positions of quinazolinone ring, having fixed with 3,5-difluoroaniline or 3,5-dichloroaniline at 2 position (Table 2). Compounds with electron-withdrawing groups, such as 7-trifluoromethyl (10a and 11a) With the compounds having potent activities toward SARS-CoV-2, we tested anti-MERS-CoV activities ( Table 3). All compounds except 10b and 11b showed good antiviral activities (IC 50 = 0.39-3.1 μM). It seems that our quinazolinone compounds are potent broad spectrum coronavirus inhibitors. In particular, all the above compounds except 10a and 11a displayed no obvious cytotoxicity (CC 50 > 25 μM).
We selected the compounds for the purpose of further pharmacological investigations on SARS-CoV-2. In the case of electronwithdrawing substituents in the aromatic ring of quinazolinone, compound 9g with chloro substituent at position 7 was selected because it has the highest SI (1 1 0). And we chose compound 11e with electrondonating substituent (hydroxyl) at 5 position because it is the most active on SARS-CoV-2 and has the highest SI (1 6 8). Compound 9g and 11e were further evaluated for their microsomal stabilities, cytotoxicities, human ether a-go-go (hERG) bindings, plasma protein bindings  (Table 4 and 5). First, the results of microsomal stability show that 9g is stable in mouse, rat, and human and 11e is microsomally more unstable than 9g, which seems to be due to the presence of -OH in the structure. In the cytotoxicity assay, 9g and 11e showed a little toxicity in HFL-1 (CC 50 = 7.5 and 9.6 μM), but did not show toxicity in the rest of the cells (Table 4).
In hERG channel inhibition assay, it was found that both compounds 9g and 11e did not interact with hERG channel. In the PPB assay, both compounds 9g and 11e showed high binding rates (97-100%). The results of CYP inhibition assay show that 9g showed little inhibition of the CYP enzyme and 11e displayed some inhibition (>65% CYP inhibition at 10 μM) in CYP1A2 and CYP3A4.
The preliminary pharmacokinetic properties of 9g and 11e were investigated by intravenous (i.v.) and oral (p.o) routes in rats, with 2 mg/ kg and 10 mg/kg, respectively ( Table 6). The oral bioavailability of 9g and 11e were 15.6% and 7.8%, respectively, partly because the clearance of compound 11e is higher than that of 9g. The PK profile of 9g seemed to be acceptable for the discovery of anti-coronavirus drugs.
In conclusion, we designed and developed 2-aminoquinazolin-4 (3H)-one derivatives as potent inhibitors against both SARS-CoV-2 and MERS-CoV. Among them, 7-chloro-2-((3,5-dichlorophenyl)amino)quinazolin-4(3H)-one (9g) and 2-( Our two lead compounds also showed good microsomal stabilities, relatively low cytotoxicities, low hERG binding affinities and CYP inhibitions. The PK profile of 9g seemed to be acceptable for the discovery of antivirals. 2-Aminoquinazolinone derivatives were found to be a promising new scaffold against coronaviruses and further optimizations to increase pharmacokinetic profiles are currently underway.

Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.