key: cord-0774524-5tndxhc0
authors: Chemboli, Raviteja; Kapavarapu, Ravikumar; Deepti, K.; Prasad, K.R.S.; Reddy, Alugubelli Gopi; Kumar, A. V. D. Nagendra; Rao, Mandava Venkata Basaveswara; Pal, Manojit
title: Pyrrolo[2,3-b]quinoxalines in attenuating cytokine storm in COVID-19: their sonochemical synthesis and in silico / in vitro assessment
date: 2021-01-03
journal: J Mol Struct
DOI: 10.1016/j.molstruc.2020.129868
sha: ff545c39b293a8813142906589fdd50379609e5b
doc_id: 774524
cord_uid: 5tndxhc0

In view of the recent global pandemic caused by COVID-19 intense efforts have been devoted worldwide towards the development of an effective treatment for this disease. Recently, PDE4 inhibitors have been suggested to attenuate the cytokine storm in COVID-19 especially tumour necrosis factor alpha (TNF-α). In our effort we have explored the 2-substituted pyrrolo[2,3-b]quinoxalines for this purpose because of their potential inhibitory properties of PDE-4 / TNF-α. Moreover, several of these compounds appeared to be promising in silico when assessed for their binding affinities via docking into the N-terminal RNA-binding domain (NTD) of N-protein of SARS-CoV-2. A rapid and one-pot synthesis of this class of molecules was achieved via the Cu-catalyzed coupling-cyclization-desulfinylation of 3-alkynyl-2-chloroquinoxalines with t-butyl sulfinamide as the ammonia surrogate under ultrasound irradiation. Most of these compounds showed good to significant inhibition of TNF-α in vitro establishing a SAR (Structure Activity Relationship) within the series. One compound e.g. 3i was identified as a promising hit for which the desirable ADME and acceptable toxicity profile was predicted in silico.

Being pathogens with very long single-strained RNA the Human Corona Viruses (HCoVs) are capable of causing a number of diseases via various pathogenic mechanisms. Indeed, due to the higher mutation rate of RNA viruses over DNA viruses they can undergo easy mutation thereby infecting various animal and human species. Among the six HCoVs such as HCoV-NL63 and HCoV-229E as Alpha-CoVs and HCoV-OC43, HCoV-HKU1, SARS-CoV and MERS-CoV as Beta-CoVs reported till 2019 end the SARS-CoV (Severe acute respiratory syndrome Corona Virus) and MERS-CoV (Middle East respiratory syndrome Corona Virus) were identified as the most aggressive HCoVs that may cause severe and fatal respiratory infections and / or multi-organ failure [1, 2] . The outbreak of SARS-CoV-1 was initially witnessed in [4] . Due to the lack of definitive vaccines and therapeutic drugs there was an urgent need in devoting efforts to identify effective therapeutics via employing diverse approaches. Accordingly, an immunomodulant as well as anti-malaria drug chloroquine (A, Fig 1) was examined and the in vitro studies [5] indicated its effectiveness against SARS-CoV and MERS-CoV [6, 7] . A close analogue of A i.e. the drug hydroxychloroquine was also explored as an experimental treatment for COVID-19 [8] . Favipiravir or T-705 (B, Fig. 1 ), the first approved drug in China [9] and explored in Japan, India etc is a pyrazine based anti-influenza drug that causes selective inhibition of RNA-dependent RNA polymerase of influenza virus [10] . The other agents that are currently in clinical trials for SARS-CoV-2 mostly include Remdesivir, Indinavir, Saquinavir, Darunavir, ASC09, Ritonavir, Lopinavir etc [11] . The SARS-CoV possess a membrane comprising of four viral proteins [e.g. spike glycoprotein (S), membrane glycoprotein (M), nucleocapsid proteins (N) and an envelope protein (E)] that play a key role during the host cell entry and viral morphogenesis and release of S-glycoprotein on the virus surface for attaching the virus to a host receptor i.e. ACE-II [12] . In 2016, the nucleocapsid (N)-RNA binding domain was explored as a target protein for the virtual screening of molecules towards the identification of potential hits against human coronavirus (CoV-OC43) [13] . Accordingly, the quinoline derivative (C, Fig. 1 ) was identified as the most promising hit and the crystallographic structure of C bound with HCoV-OC43 was deposited. Recently, a crystallographic structure of Apo-nucleocapsid (RNA-binding domain) of SARS-CoV-2/COVID-19 (PDB: 6M3M) has revealed the potential drug targeting sites [14] . In earlier study using HCoV-OC43 as a model for CoV (in 2014) Hou et al presented the 3D structure of HCoV-OC43 Nterminal domain (N-NTD) complexed with ribonucleoside 5′-monophosphates to identify a distinct ribonucleotide-binding pocket [15] . Indeed, a new inhibitor PJ34 (D, Fig. 1 ) was identified by targeting this pocket through virtual screening and the crystal structure of the N-NTD-PJ34 complex was determined.

Nevertheless, efforts towards identification of small organic molecule based inhibitor against nucleocapsid (N) of SARS-CoV-2 is rather uncommon. In our earlier effort we have explored alkynyl substituted chloroquinoxaline derivatives E (Fig. 2 ) as potential ligands for N-protein of SARS-CoV-2 in silico [16] . In further continuation of this research we became interested in exploring the related 2-substituted pyrrolo [2,3-b] quinoxalines F (Fig. 2 ) as potential agents for COVID-19. Notably, several of these compounds showed high inhibition (> 90% at 30 µM) of PDE4B (phosphodiesterase 4B) in vitro previously though the assay results remained inconclusive due to the interference caused by the direct luciferase inhibition of these compounds [17] . Recently the PDE4 inhibitors, known treatment for asthma, chronic obstructive pulmonary disease (COPD), psoriasis etc have been predicted to be a valuable therapeutic option for COVID-19 ( Fig. 2 ) due to their upstream inhibition of multiple cytokine signaling pathways along with the regulation of the pro-inflammatory/anti-inflammatory balance [18] . Indeed, because of targeting an early stage inflammatory response and ameliorating lung inflammation, PDE4 inhibitors might be effective for the initial phase of COVID-19 pneumonia prior to the onset of cytokine storm and severe multi-organ dysfunction [19] . 

To validate our predictive rational in favor of selecting the template F the related docking studies were carried out in silico. The iGEMDOCK version 2.1 software [20] (a graphical automatic drug design system for docking, screening and analysis) was used for this purpose. It is a program for computing ligand conformation and orientation relative to the active site of the protein. The in silico docking simulation studies were performed to evaluate the molecular interactions of a series of compounds e.g. 3a-j derived from F along with the reference compound C (Fig 1) with the SARS-CoV-2 nucleocapsid protein N-terminal RNA binding domain (PDB: 6M3M). The corresponding binding energies and the interaction diagrams are presented in Table 1 and 2, respectively. Notably, the results obtained via iGEMDOCK were further validated using the two other docking programs such as DOCKTHOR [21] and SWISSDOCK [22] ( Table 1) . Among the compounds examined (Table 1 ) the molecule 3c, 3d and 3j showed superior binding energies across all three docking programs i.e. GEMDOCK, DOCKTHOR and SWISSDOCK used whereas 3i and 3g showed encouraging results in case of GEMDOCK and DOCKTHOR. Moreover, the interaction diagrams (Table 2) (Table 3) . Indeed, the nitrogen of the central fused ring was involved in the H-bonding interaction with TRP80 in all these cases whereas - interactions was observed with TYR71 and LYS75 in most of the cases. The 2D and 3D interaction diagram of compound 3c and 3i with SARS-CoV-2 nucleocapsid [23, 24, 25] include the Pd-mediated intramolecular ring closure of 2-alkynyl-3trifluoroacetamidoquinoxalines [24] or action of amines on 2-chloro-3-alkynylquinoxalines [25] . In our earlier effort the 2-substituted pyrrolo[2,3-b]quinoxalines having free NH were prepared directly via the Cu-catalyzed coupling-cyclization of 3-alkynyl-2-chloroquinoxalines with methanesulfonamide in a single pot [17] . While the methodology afforded the desired products in good yields the duration of the reaction was 4-8h. It was therefore desirable to establish a faster access to this class of compounds. The use of ultrasound irradiation is a well-known option in accelerating the reaction rate considerably thereby reducing the reaction time significantly. It is therefore not surprising that the ultrasound irradiation being a useful alternative to the conventional energy sources such as heat, light, or ionizing radiation has found wide applications in organic synthesis [26] [27] [28] [29] . Indeed, the ultrasound-assisted reactions have become common approaches in the area of green and sustainable chemistry [26] because of (i) their environmentally friendly nature (via minimization of waste generation), [27] (ii) decreased energy requirements [28] and (iii) increased efficiency and effectiveness. Thus we became interested in adopting this approach in our current effort.

Initially, we examined the reaction of 2-chloro-3-(phenylethynyl)quinoxaline (1a) with methanesulfonamide (NH 2 SO 2 CH 3 ) (2a) under ultrasound irradiation at 80 °C without altering the other reaction conditions used earlier [17] . The reaction was carried out in the presence of Cu(OAc) 2 and Et 3 N in DMF using a laboratory ultrasonic bath SONOREX SUPER RK 510H model producing irradiation of 35 kHz (entry 1, Table 4 ). While the reaction was completed within 2h the yield of desired product 3c was not particularly high. The yield was not improved when 2a was replaced by other sulfonamides such as benzene sulfonamide (2b) (entry 2, Table 4 ) or p-toluene sulfonamide (2c) (entry 3, Table 4 ). Notably, the product yield was increased considerably when t-butyl sulfinamide (racemic) (2d) was used (entry 4, Table   4 ) and the reaction was completed within 1.5 h. Encouraged by this observation we continued the optimization studies. However, change of base from Et 3 N to DBU decreased to product yield (entry 5, Table 4 ) whereas increase of reaction time did not improve the yield significantly (entry 6, Table 4 ). The reaction was also performed under silent conditions (entry 7, Table 4 ). While the coupling-cyclization proceeded under this condition the product 3c was isolated in inferior yield even after 4h. All these reactions were carried out using slight excess of stoichiometric amount of 2. In order to understand the role of sulfinamide the reaction was carried out using half equivalent of 2d when the reaction did not reach to the completion affording 3c in inferior yield (entry 8, Table 4 ). Further, these reactions were catalyzed by 1 mol% of Cu(OAc) 2 whereas decrease in catalyst quantity was found to be counterproductive (entry 9, Table   4 ). Similarly, the effect of quantity of base i.e. 1.5 equivalent of Et 3 N used was also examined and was found to have negative impact on the product yield when used in decreased quantity i.e. 1 equivalent or less. Notably, while the current coupling-cyclization-desulfinylation was found to be insensitive to the presence of air however maintaining the anhydrous atmosphere was necessary in order to avoid the partial hydrolysis of the starting chloro derivative (1a) as a side reaction. Thus the reaction does not require the use of any closed reaction vessel thereby avoiding the pressure development inside the vessel.

Nevertheless, the condition of entry 4 of Table 1 appeared to be optimum for the preparation of 3c under ultrasound irradiation. The reaction was performed using 1.4 mmol of 2d. f The reaction was performed using 0.014 mmol of Cu(OAc) 2 . The optimized reaction conditions were then used to prepare all the target compounds 3c-j (Table 1) and results are presented in Table 5 . A range of 2-chloro-3-alkynyl quinoxaline derivatives (1) were employed for this purpose and the ultrasound assisted reaction proceeded well in all these cases affording the corresponding 2-substituted pyrrolo [2,3-b] quinoxalines (3c-j) in good to acceptable yield. Notably and expectedly, unlike the previous report [17] the use of 2d did not afford the corresponding 1-alkylsulfonyl substituted analogues such as 3a and 3b (Table 1 ) under the conditions employed. Thus access of these compounds were made via the earlier non-sonochemical method [17] . Nevertheless, all the synthesized pyrrolo[2,3-b]quinoxaline derivatives (3) were characterized by the spectral ( 1 H and 13 C NMR and Mass) data. The partial 1 H and 13 C NMR data of a representative compound 3e is presented in Fig 4. It is evident that the methyl protons of t-butyl group appeared at 1.46 in the 1 H NMR spectra whereas the C-3 proton and NH of the pyrrolo [2,3-b] quinoxaline ring appeared at 6.65 and 9.13 δ respectively. The methyl carbons of t-butyl group appeared at 31.4 ppm in the 13 C NMR spectra whereas the carbon bearing all these Me groups appeared at 40.5 ppm. The 13 C NMR signals at 99.4 and 150.9 ppm were due to the C-3 and one of the ring junction carbon as shown in Fig 4, respectively. The presence of NH moiety was further supported by the IR absorption near 3200 cm -1 . Fig. 4 . Partial representation of 1 H and 13 C NMR spectral data of compound 3e.

Based on results of Table 2 and previous reports [17, 30] In view of the fact that inhibition of PDE4 may attenuate the cytokine storm in COVID-19 (through the upstream inhibition of pro-inflammatory molecules) particularly tumour necrosis factor alpha (TNF-α) [19] we examined the TNF- inhibitory potential of all the synthesized compounds (3a-j) in vitro. All the compounds were tested initially at 10 μM concentration and the known TNF- inhibitors rolipram [32] and thalidomide [33] was used as reference compounds in this assay (Table 6 ). While most of the compounds showed good to significant inhibition of TNF- the compound 3c, 3d, 3i and 3j appeared to be promising (inhibition > 70%) among them. Besides, 3a, 3b, 3f and 3g were also found to be effective (inhibition > 60%). The nature and size of the substituent attached to the pyrrolo [2,3-b] quinoxaline ring at C-2 position appeared to have played a key role in the inhibition of TNF- (Fig 5) . The effectiveness of substituents was found to be in the order aryl moiety > long chain alkyl / hydroxyl alkyl > H / t-butyl. While the "Me" group was found be marginally better than "H" at C-5 position however activity was decreased when "H" was replaced by "MeSO 2 " group at N-1 position. Nevertheless, except 3e and 3h (inhibition < 60%) the IC 50 values of rest of the compounds were determined (Table 6 ). Accordingly, the compound 3i with IC 50 ~ 5.14 ± 0.34 µM (Fig. 6 ) was identified as the most potent inhibitor among them. While 3i was not better than the reference compound rolipram however it was found to be superior to another known inhibitor thalidomide (Table 6 ). Next, an initial assessment about ADME (absorption, distribution, metabolism, and excretion) or pharmacokinetic properties of compound 3i along with the two other compounds 3c and 3j was carried out using Swiss ADME web-tool [34] and results are summarized in Table 7 (among the various descriptors predicted for these compounds. Notably, none of these compounds showed violation of Lipinski or Veber rule. Finally, to gain some initial idea about potential toxicity of the identified hit 3i the toxicity prediction of this compound was carried out using the pkCSM web-tool [35] (Table 8) . Though the predicted AMES toxicity (mutagenicity or genetic toxicity) could be a possible concern the molecule was predicted to be safe in terms of hepatotoxicity, hERG I (human ether-a-go-go) inhibition, skin sensitisation etc. The molecule also appeared to be safe from the view point of predicted value of oral rat acute and chronic toxicity. Thus the molecule 3i may have medicinal value and deserves further evaluation towards the identification of prospective agent against COVID-19.

In conclusion, we have explored the 2-substituted pyrrolo [2,3-b] quinoxalines as potential agents against COVID-19. Several of these compounds appeared to be promising in silico when assessed for their binding affinities via docking into the N-terminal RNA-binding domain (NTD) of N-protein of SARS-CoV-2. A rapid synthesis of this class of molecules was achieved via the Cu-catalyzed coupling-cyclizationdesulfinylation of 3-alkynyl-2-chloroquinoxalines with t-butyl sulfinamide as the ammonia surrogate under ultrasound irradiation. This one-pot method afforded the desired products in good yield, better than that of the previously reported method. Two N-sulfonyl analogues were also accessed via the reported method.

Most of these compounds showed good to significant inhibition of TNF- in vitro establishing a SAR (Structure Activity Relationship) within the series. In general the N-sulfonyl analogues were found to be less effective than the pyrrolo[2,3-b]quinoxalines having free NH group. One of the three best active compounds i.e. 3i was identified as a promising hit for which the desirable ADME and acceptable toxicity profile was predicted in silico. The compound 3i deserves further evaluation towards the identification of a prospective agent against COVID-19.

Research and Development on Therapeutic Agents and Vaccines for COVID-19 and Related Human Coronavirus Diseases

SARS and MERS: recent insights into emerging coronaviruses

A pneumonia outbreak associated with a new coronavirus of probable bat origin

Remdesivir and chloroquine effectively inhibit the recently emerged novel coronavirus (2019-nCoV) in vitro

Effects of chloroquine on viral infections: an old drug against today's diseases?

Chloroquine for the 2019 novel coronavirus SARS-CoV-2

A systematic review on the efficacy and safety of chloroquine for the treatment of COVID-19

Anti-HCV, nucleotide inhibitors, repurposing against COVID-19

Favipiravir (T-705), a novel viral RNA polymerase inhibitor

Coronavirus puts drug repurposing on the fast track

Anti-SARS coronavirus agents: a patent review (2008 -present)

Structure-based virtual screening and experimental validation of the discovery of inhibitors targeted towards the human coronavirus nucleocapsid protein

6M3M: Structural insights of SARS-CoV-2 nucleocapsid protein RNA binding domain reveal potential unique drug targeting sites

Ultrasound assisted synthesis of 3-alkynyl substituted 2-chloroquinoxaline derivatives

Pyrrole[2,3-b]quinoxalines as inhibitors of firefly luciferase: Their Cu-mediated synthesis and evaluation as false positives in a reporter Gene Assay

Rationale for Evaluating PDE4 Inhibition for Mitigating against Severe Inflammation in COVID-19 Pneumonia and Beyond

Commentary: Phosphodiesterase 4 inhibitors as potential adjunct treatment targeting the cytokine storm in COVID-19

GEMDOCK: A generic evolutionary method for molecular docking

Genetic Evolutionary Method for molecular Docking

Dock Thor: A receptor-ligand docking program

SwissDock: a web service to predict the molecular interactions that may occur between a target protein and a small molecule

XXI : Synthesis of 2-Substituted Furo [2, 3-b] quinoxaline and 2-Substituted Pyrrolo [2, 3-b] quinoxaline

3-Disubstituted pyrrolo[2,3-b]quinoxalines via aminopalladation-reductive elimination

Alkynyl-and dialkynyl-quinoxalines. Synthesis of condensed quinoxalines

Ultrasound in synthetic organic chemistry

Ultrasonic Reactions, in Green Techniques for Organic Synthesis and Medicinal Chemistry

Recent Advances in the Ultrasound-Assisted Synthesis of Azoles

Applications of Ultrasound in Organic Synthesis -A Green Approach

Efficient indoles and anilines syntheses employing tert-butyl sulfinamide as ammonia surrogate

For complexation of S=O group with Cu(OAc) 2 see: E. F. Weaver, W. Keim, A preliminary study of differential thermal analysis, conductance and cryoscopic behavior of dimethyl sulfoxide complexes

Characterization of cAMP-dependent inhibition of LPS induced TNF a production by rolipram, a specific phosphodiesterase IV (PDE IV) inhibitor

Thalidomide exerts its inhibitory action on tumor necrosis factor alpha by enhancing mRNA degradation

SwissADME: a free web tool to evaluate pharmacokinetics, drug-likeness and medicinal chemistry friendliness of small molecules

pkCSM: predicting small-molecule pharmacokinetic properties using graph-based signatures

Kumar were involved in the preparation, isolation, purification and characterization of all the target compounds presented in the current manuscript

Ravikumar Kapavarapu and Alugubelli Gopi Reddy were involved in performing all the in silico studies as well as in vitro assays

The authors thank the management of Dr. Reddy's Institute of Life Sciences, Hyderabad, India, for continuous support and encouragement.

Supplementary data to this article can be found online at xxxxxxxxx